CN117019889A - Method and system for intelligently detecting faults of three-roller rotary rolling equipment - Google Patents
Method and system for intelligently detecting faults of three-roller rotary rolling equipment Download PDFInfo
- Publication number
- CN117019889A CN117019889A CN202310886772.9A CN202310886772A CN117019889A CN 117019889 A CN117019889 A CN 117019889A CN 202310886772 A CN202310886772 A CN 202310886772A CN 117019889 A CN117019889 A CN 117019889A
- Authority
- CN
- China
- Prior art keywords
- information
- dynamic
- rolling equipment
- rotary rolling
- roller rotary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005096 rolling process Methods 0.000 title claims abstract description 123
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001514 detection method Methods 0.000 claims abstract description 94
- 230000004927 fusion Effects 0.000 claims abstract description 80
- 230000003068 static effect Effects 0.000 claims abstract description 72
- 238000012423 maintenance Methods 0.000 claims description 46
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 238000005461 lubrication Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 7
- 230000010354 integration Effects 0.000 claims description 7
- 238000009987 spinning Methods 0.000 claims 1
- 238000007405 data analysis Methods 0.000 abstract description 5
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 abstract description 2
- 229910052802 copper Inorganic materials 0.000 abstract description 2
- 239000010949 copper Substances 0.000 abstract description 2
- 238000004140 cleaning Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 239000002184 metal Substances 0.000 description 5
- 229910052751 metal Inorganic materials 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 239000010687 lubricating oil Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012847 principal component analysis method Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B37/00—Control devices or methods specially adapted for metal-rolling mills or the work produced thereby
- B21B37/78—Control of tube rolling
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The application relates to the technical field of copper pipe rolling, in particular to a method and a system for intelligently detecting faults of three-roller rotary rolling equipment, wherein the method comprises the following steps: collecting equipment parameters of three-roller rotary rolling equipment in a power-off state, and obtaining static multielement information; determining static standardized information of the three-roller rotary rolling equipment, and judging whether the static multi-element information meets the static standardized information; collecting parameter information of the three-roller rotary rolling equipment in the working state, and obtaining dynamic multielement information; feature fusion is carried out on feature data in the dynamic multielement information to obtain dynamic feature fusion information; establishing a dynamic detection model, inputting dynamic characteristic fusion information, and obtaining a fault detection result; and overhauling the three-roller rotary rolling equipment through a fault detection result. The application effectively solves the problems of high labor cost, low efficiency, strong subjectivity and the like of the traditional method. The accuracy and the efficiency of fault detection are improved, automatic detection is realized, and more comprehensive data analysis and decision support are provided.
Description
Technical Field
The application relates to the technical field of copper pipe rolling, in particular to a method and a system for intelligently detecting faults of three-roller rotary rolling equipment.
Background
The three-roller rotary rolling equipment is an important equipment for metal processing and is widely applied to industries such as steel, nonferrous metals and the like. The metal blank is rolled for a plurality of times, so that the shape and the size of the metal blank are processed, and the metal blank has the characteristics of high efficiency, accuracy and reliability. However, due to long operation and high load operation of the equipment, risks of failure and damage may be faced, resulting in production interruption, quality problems and safety hazards.
The traditional fault detection method of the three-roller rotary rolling equipment mainly relies on manual inspection and experience judgment, and has the problems of high labor cost, low efficiency, strong subjectivity and the like. In addition, the traditional method has limited analysis capability on complex fault modes and multidimensional parameters, and the type and severity of the fault are difficult to accurately judge.
The information disclosed in this background section is only for enhancement of understanding of the general background of the disclosure and is not to be taken as an admission or any form of suggestion that this information forms the prior art that is well known to a person skilled in the art.
Disclosure of Invention
The application provides a method and a system for intelligently detecting faults of three-roller rotary rolling equipment, which can effectively solve the problems in the background technology.
In order to achieve the above purpose, the technical scheme adopted by the application is as follows:
a method for intelligently detecting faults of three-roller rotary rolling equipment, the method comprising:
collecting equipment parameters of three-roller rotary rolling equipment in a power-off state, and obtaining static multielement information;
determining static standardized information of the three-roller rotary rolling equipment, judging whether the static multi-element information meets the static standardized information, if not, outputting the static multi-element information, and if so, executing the next step;
collecting parameter information of the three-roller rotary rolling equipment in the working state, and obtaining dynamic multielement information;
feature fusion is carried out on the feature data in the dynamic multivariate information to obtain dynamic feature fusion information;
establishing a dynamic detection model, and inputting the dynamic characteristic fusion information into the dynamic detection model to obtain a fault detection result;
and overhauling the three-roller rotary rolling equipment according to the fault detection result.
Further, the collecting the equipment parameters of the three-roller rotary rolling equipment in the power-off state, and obtaining the static multielement information comprises the following steps:
cleaning the three-roller rotary rolling equipment;
the data of each position of the processed three-roller rotary rolling equipment is acquired, and a static data subset is obtained by taking each position as a unit;
the lubrication detection is carried out on the three-roller rotary rolling equipment, and a lubrication detection result is obtained;
and obtaining the static multivariate information through the static data subsets of the positions and the lubrication detection results.
Further, the collecting parameter information under the working state of the three-roller rotary rolling device and obtaining dynamic multi-element information comprise:
cleaning the three-roller rotary rolling equipment;
starting the three-roller rotary rolling equipment, and collecting multi-dimensional parameters of the three-roller rotary rolling equipment in a working state, wherein the multi-dimensional parameters can reflect the dynamic related state of the three-roller rotary rolling equipment;
preprocessing the multi-dimensional parameters, and extracting dynamic characteristic information from the preprocessed multi-dimensional parameters;
carrying out recognition calculation on the dynamic characteristic information and obtaining effective characteristic judgment information;
and obtaining the dynamic multivariate information through the preprocessed multidimensional parameters and the effective characteristic judgment information.
Further, the performing feature fusion on the feature data in the dynamic multivariate information to obtain dynamic feature fusion information includes:
according to the effective characteristic judgment information corresponding to the data type in the multidimensional parameter, the identification degree of each effective characteristic judgment information to each position of the three-roller rolling equipment is obtained;
and carrying out feature fusion on the identification degree of each piece of corresponding effective feature judgment information, and carrying out correlation integration on different features to obtain the dynamic feature fusion information.
Further, the obtaining the identification degree of each effective feature judgment information to each position of the three-roller rolling equipment includes:
the historical maintenance record of the three-roller rotary rolling equipment is called;
counting data characteristics in a plurality of data types according to the fault reasons in the historical maintenance records;
and the fault reasons correspond to data characteristics in a plurality of data types of the three-roller rotary rolling equipment and are mapped to the identification degree of the effective characteristic judgment information on each position of the three-roller rotary rolling equipment.
Further, the feature fusion is performed on the recognition degrees of the effective feature judgment information, and the correlation integration is performed between different features to obtain the dynamic feature fusion information, which includes:
the feature fusion adopts a weighted fusion algorithm, and the weighted fusion algorithm is as follows:
W＝a1*w1+a2*w2+……+ai*wi
wherein W is characteristic fusion information of one position of the three-roller rotary rolling equipment; a1 and a2 … … ai are the recognition degrees of the data types of various types to the position; w1 and w2 … … wi are respectively corresponding to the effective characteristic judgment information;
and collecting and integrating the characteristic fusion information of each position of the three-roller rotary rolling equipment to obtain the dynamic characteristic fusion information.
Further, the establishing the dynamic detection model includes:
collecting historical detection maintenance information, and preprocessing the historical detection maintenance information;
formulating a fault maintenance type according to the preprocessed historical detection maintenance information, wherein the fault maintenance type comprises corresponding fault parameter information;
and selecting a proper learning algorithm to learn the fault maintenance type, and establishing a dynamic detection model.
Further, inputting the dynamic feature fusion information into the dynamic detection model to obtain a fault detection result, including:
the dynamic characteristic fusion information is matched with the fault maintenance type in the dynamic detection model, and a dynamic matching result is obtained;
setting a dynamic matching interval, judging whether the dynamic matching result is in the dynamic matching interval, and if so, acquiring the fault detection result according to the corresponding fault maintenance type; if not, the fault is marked as an in-doubt fault.
A system for intelligent fault detection of a three-roll rotary rolling apparatus, the system comprising:
the static information acquisition module is used for acquiring equipment parameters of the three-roller rotary rolling equipment in a power-off state and acquiring static multielement information;
the static standard judging module is used for determining static standardized information of the three-roller rotary rolling equipment, judging whether the static multi-element information meets the static standardized information, outputting the static multi-element information if the static multi-element information does not meet the static standardized information, and executing the next step if the static multi-element information meets the static standardized information;
the dynamic information acquisition module is used for acquiring parameter information under the working state of the three-roller rotary rolling equipment and acquiring dynamic multielement information;
the feature information fusion module is used for carrying out feature fusion on the feature data in the dynamic multivariate information to obtain dynamic feature fusion information;
the fault detection module is used for establishing a dynamic detection model, inputting the dynamic characteristic fusion information into the dynamic detection model and obtaining a fault detection result; and overhauling the three-roller rotary rolling equipment according to the fault detection result.
Further, the dynamic information acquisition module includes:
the parameter acquisition unit is used for starting the three-roller rotary rolling equipment and acquiring multi-dimensional parameters of the three-roller rotary rolling equipment in a working state, wherein the multi-dimensional parameters can reflect the dynamic relevant state of the three-roller rotary rolling equipment;
the feature extraction unit is used for preprocessing the multi-dimensional parameters and extracting dynamic feature information from the preprocessed multi-dimensional parameters;
the identification calculation unit is used for carrying out identification calculation on the dynamic characteristic information and obtaining effective characteristic judgment information;
and the output unit is used for obtaining the dynamic multivariate information through the preprocessed multidimensional parameter and the effective characteristic judgment information.
By the technical scheme of the application, the following technical effects can be realized:
the problems of high labor cost, low efficiency, strong subjectivity and the like in the traditional method are effectively solved. The accuracy and the efficiency of fault detection are improved, automatic detection is realized, and more comprehensive data analysis and decision support are provided.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings may be obtained according to the drawings without inventive effort to those skilled in the art.
FIG. 1 is a flow diagram of a method for intelligently detecting faults in a three-roll rotary rolling device;
FIG. 2 is a schematic flow chart of obtaining static multivariate information;
FIG. 3 is a schematic flow chart of obtaining dynamic multivariate information;
FIG. 4 is a flow chart of obtaining the identification degree of the device by the effective feature judgment information;
fig. 5 is a schematic structural diagram of a system for intelligently detecting faults of a three-roller rotary rolling device.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Example 1
As shown in fig. 1, the application provides a method for intelligently detecting faults of three-roller rotary rolling equipment, which comprises the following steps:
s100: collecting equipment parameters of three-roller rotary rolling equipment in a power-off state, and obtaining static multielement information;
in particular, suitable sensors or measuring instruments are selected to accurately identify and collect the required device parameters. And the data acquisition equipment is correctly configured and arranged, so that the data acquisition equipment can work normally, and the equipment parameter data is recorded and stored. In addition, it is ensured that the data acquisition device is properly connected to the computer or data storage device. In the power-off state, the data acquisition time is ensured to be long enough to acquire sufficient data samples. Meanwhile, the accuracy of the sensor is checked, and the influence of errors and drift on acquired data is avoided. The collected equipment parameter data is converted into static multivariate information through a proper feature extraction and data analysis method, and proper feature extraction algorithm and data processing technology are selected to capture key features of equipment states, so that reliable basic data is provided for subsequent fault detection and analysis.
S200: determining static standardized information of the three-roller rotary rolling equipment, judging whether the static multi-element information meets the static standardized information, if not, outputting the static multi-element information, and if so, executing the next step;
specifically, according to design specifications, production requirements or historical data and the like of the three-roller rotary rolling equipment, standardized information in a static working state of the equipment is determined, wherein the standardized information comprises upper and lower limits, a variation range, a reasonable interval and the like of parameters, the acquired equipment parameter data in a power-off state of the three-roller rotary rolling equipment is compared and analyzed with the determined static standardized information, and whether the static standardized information is met is determined. If the static multielement information accords with the static standardized information, the equipment is in a normal running state, and the subsequent steps can be continued; if not, static multivariate information needs to be output to indicate that an abnormality or potential failure exists in the device.
S300: collecting parameter information of the three-roller rotary rolling equipment in the working state, and obtaining dynamic multielement information;
specifically, a proper sensor or measuring instrument is selected and correctly installed on the three-roller rotary rolling equipment, so that the required parameter data can be accurately and stably acquired. The data acquisition device is configured to be connected to a computer or data storage device. Under the working state of the three-roller rotary rolling equipment, starting the data acquisition equipment, acquiring parameter information, acquiring real-time data of key parameters of the equipment through a sensor, storing the acquired parameter data into a database or a file, performing necessary data cleaning and preprocessing, and ensuring the accuracy and the integrity of the data. And carrying out feature extraction and data analysis based on the acquired parameter data to obtain dynamic multivariate information, wherein the information comprises the variation trend, the fluctuation, the correlation and the like of the parameters.
S400: feature fusion is carried out on feature data in the dynamic multielement information to obtain dynamic feature fusion information;
specifically, each parameter or feature is evaluated and selected based on the acquired dynamic multivariate information, and the most representative and distinguishing feature is selected according to the importance, correlation, distinguishing degree and other indexes of the feature. The selected features are further extracted and transformed using appropriate algorithms and methods, for example, statistical features, frequency domain features, wavelet transforms, etc. techniques may be used to capture the relationships between the parameters and time-varying characteristics. And fusing the extracted characteristic data to generate dynamic characteristic fusion information. The fusion mode can comprise simple methods of weighted summation, principal component analysis and the like, so that comprehensive characteristics can be ensured to better describe the working state of the equipment. And carrying out normalization processing on the fused dynamic characteristics to eliminate the influence brought by different characteristic dimensions and enable the characteristic values to be in the same range so as to be further analyzed and used later.
S500: establishing a dynamic detection model, inputting dynamic characteristic fusion information into the dynamic detection model, and obtaining a fault detection result;
specifically, the collected dynamic feature fusion information is used as input data, and corresponding fault states (normal or abnormal) are marked at the same time. The data is divided into training and testing sets. Cross-validation or time series methods are typically employed to ensure that the model has a good generalization ability. The proper machine learning or deep learning algorithm is selected as a dynamic detection model, for example, a support vector machine, a random forest neural network and the like can be used, and the model can be applied only by achieving the effect of fault detection on the three-roller rolling equipment. Further feature processing and preprocessing, such as normalization, dimension reduction, feature selection, etc., are performed on the input data to improve model performance and effectiveness. Training the selected model by using a training set, and adjusting the model super-parameters and the optimization strategy to achieve better fault detection performance. The test set is used to evaluate performance metrics of the model, such as accuracy, recall, precision, etc., to evaluate the failure detection effect of the model. And applying the trained dynamic detection model to actual three-roller rotary rolling equipment, performing fault detection by using the acquired real-time data input model, and judging whether the equipment has an abnormality or potential fault according to the result output by the model.
S600: and overhauling the three-roller rotary rolling equipment through a fault detection result.
Specifically, staff knows the type and the reason of the fault through the fault detection result, can directly maintain and replace the part with the fault, and the equipment is guaranteed to be put into use fast.
By the technical scheme, the problems of high labor cost, low efficiency, strong subjectivity and the like in the traditional method are effectively solved. The accuracy and the efficiency of fault detection are improved, automatic detection is realized, and more comprehensive data analysis and decision support are provided.
Further, as shown in fig. 2, collecting the equipment parameters of the three-roller rolling equipment in the power-off state, and obtaining the static multivariate information includes:
s110: cleaning the three-roller rotary rolling equipment;
s120: the data of each position of the processed three-roller rotary rolling equipment is acquired, and a static data subset is obtained by taking each position as a unit;
s130: lubrication detection is carried out on the three-roller rotary rolling equipment, and a lubrication detection result is obtained;
s140: and obtaining static multivariate information through the static data subsets of the positions and the lubrication detection results.
Specifically, before collection, the surface of the equipment is cleaned, so that the surface of the three-roller rotary rolling equipment is ensured to be clean, and impurities and dirt which can affect the data collection quality are removed. And selecting proper sensors, instruments or data acquisition equipment, determining a proper acquisition method to divide the three-roller rotary rolling equipment into various positions or areas according to the characteristics and requirements of equipment parameters, and determining the parameters to be acquired at each position according to the equipment structure, the working principle and key components. And sequentially carrying out data acquisition on each position, integrating the data of the same position into the same static data subset, constructing different subsets according to the divided positions or areas, detecting the lubrication condition of the three-roller rotary rolling equipment by using a proper method and an instrument, for example, acquiring a sample of lubricating oil for chemical analysis, or monitoring parameters such as temperature, viscosity and the like of the lubricating oil, and carrying out lubrication detection so as to monitor the lubrication state of the equipment in real time and carry out fault early warning. And (3) finishing and calculating the data by using a data processing and analyzing tool, and synthesizing a static data subset and a lubrication detection result of each position to generate static diversified information, wherein the static diversified information comprises correlation among parameters, deviation ranges, characteristic statistical values and the like.
Further, as shown in fig. 3, collecting parameter information under the working state of the three-roller rolling equipment, and obtaining dynamic multi-component information includes:
s310: cleaning the three-roller rotary rolling equipment;
s320: starting the three-roller rotary rolling equipment, and collecting multidimensional parameters of the three-roller rotary rolling equipment in a working state, wherein the multidimensional parameters can reflect the dynamic related state of the three-roller rotary rolling equipment;
s330: preprocessing the multidimensional parameter, and extracting dynamic characteristic information from the preprocessed multidimensional parameter;
s340: carrying out recognition calculation on the dynamic characteristic information and obtaining effective characteristic judgment information;
s350: and obtaining dynamic multielement information through the preprocessed multidimensional parameters and the effective characteristic judgment information.
Specifically, before collection, the surface of the three-roller rotary rolling equipment is ensured to be clean, and impurities and dirt which can affect the data collection quality are removed. In the running state of the equipment, proper sensors, instruments or data acquisition equipment are used for acquiring parameters related to the state of the equipment, such as rolling force, process temperature, vibration frequency, current, speed and the like, and preprocessing the acquired data, such as noise removal, smoothing and data normalization, is carried out, so that the reliability and the accuracy of the data are improved, and preparation is made for subsequent feature extraction and analysis. Dynamic characteristic information is extracted from the preprocessed multidimensional parameters, and a proper characteristic extraction method, such as time domain characteristics, frequency domain characteristics, statistical characteristics and the like, is selected according to the characteristics and the working state of the equipment, wherein the characteristics can reflect the change trend of the vibration frequency of the equipment, the fluctuation degree of current and other relevant states. And identifying and calculating the extracted dynamic characteristic information. And establishing a proper model or algorithm by using methods such as machine learning, pattern recognition, data mining and the like, and analyzing and judging the dynamic characteristics. This may include fault diagnosis, condition monitoring, anomaly detection, etc.
Further, feature fusion is performed on feature data in the dynamic multivariate information, and obtaining dynamic feature fusion information includes:
s410: according to the effective characteristic judgment information corresponding to the data types in the multidimensional parameters, the identification degree of each effective characteristic judgment information to each position of the three-roller rotary rolling equipment is obtained;
specifically, effective feature judgment information related to each multi-dimensional parameter is determined according to the data type of the parameter, for example: the pressure data can reflect the position of the feeding side, the position of the discharging side and the pressure states of the working roller and the supporting roller according to the detection position of the setting sensor, so that the fault condition is judged; the collection of some data may be affected by a plurality of positions, for example, the collection of vibration information in the working state may identify faults in working areas where the working roller, the supporting roller and the like apply force or generate vibration to a high degree, the identification degree is reduced for some positions where the faults cannot be reflected by vibration, and the identification degree is moderate for some positions where the vibration information can only reflect a part of the fault rate.
S420: and carrying out feature fusion on the identification degree of each corresponding effective feature judgment information, and carrying out correlation integration on different features to obtain dynamic feature fusion information.
Specifically, the recognition degree of each effective feature judgment information is subjected to feature fusion, in the feature fusion process, the correlation among different features is also required to be considered, the correlation degree among different features is evaluated by calculating the correlation coefficient among the features, covariance matrix or using a principal component analysis method and the like, redundant information can be effectively avoided, the accuracy of feature fusion is improved, and for example, whether a working roll has faults or not can be more comprehensively and accurately reflected through a pressure test, a vibration test and a thermal test; the final dynamic characteristic fusion information is obtained through the characteristic fusion and integration steps, and the information integrates a plurality of effective characteristic judgment information and the recognition degrees of the effective characteristic judgment information on different equipment positions, so that comprehensive and accurate dynamic characteristic information can be provided, and the state change and abnormal conditions of the three-roller rotary rolling equipment can be better understood.
Further, as shown in fig. 4, obtaining the recognition degree of each effective feature judgment information on each position of the three-roller rolling device includes:
s411: the historical maintenance record of the three-roller rotary rolling equipment is called;
s412: counting data characteristics in various data types according to fault reasons in the historical maintenance records;
s413: the fault reasons correspond to the data characteristics in the multiple data types and are mapped to the recognition degree of the effective characteristic judgment information on each position of the three-roller rolling equipment.
Specifically, historical maintenance records of the three-roller rotary rolling equipment are collected and arranged, and the historical maintenance records comprise information such as fault types, maintenance reasons, maintenance time and the like. The cause of the fault in the historical maintenance record is analyzed and various data types are involved, such as parameters of temperature, vibration, current and the like. The feature expression of different fault reasons on different data types is counted, for example, a certain fault reason may cause temperature rise, vibration frequency change and the like. For each failure cause, corresponding effective feature judgment information is defined according to the data features in the historical maintenance record. The recognition degree of different fault causes at different positions is evaluated by analyzing the historical data and the actually measured data, and based on the recognition degree, weights are distributed to each position and the corresponding effective characteristic judgment information to reflect the importance of the information.
Further, feature fusion is performed on the recognition degree of each corresponding effective feature judgment information, and correlation integration is performed between different features to obtain dynamic feature fusion information, which comprises the following steps:
s421: the feature fusion adopts a weighted fusion algorithm, and the weighted fusion algorithm is as follows:
W＝a1*w1+a2*w2+……+ai*wi
wherein W is characteristic fusion information of one position of the three-roller rotary rolling equipment; a1 and a2 … … ai are the recognition degrees of the data types of various types to the position; w1 and w2 … … wi are corresponding effective feature judgment information, and a1+a … … +ai=1;
s422: and collecting and integrating the characteristic fusion information of each position of the three-roller rotary rolling equipment to obtain dynamic characteristic fusion information.
Specifically, for each location, it is ensured that the degree of recognition of the location by various data types has been evaluated, and corresponding valid feature judgment information is determined for each type. The weight coefficient is set and is determined according to the specific requirements and the complexity of the problem. And carrying out feature fusion calculation by using a weighted fusion algorithm to obtain feature fusion information on each position. In the feature fusion process, if there is a correlation between different features, a statistical analysis method, such as a correlation coefficient, a covariance matrix, or principal component analysis, may be used to evaluate the degree of correlation between different features. And integrating different features according to the related analysis results, eliminating redundant information, keeping meaningful features, and collecting and integrating feature fusion information on each position to obtain dynamic feature fusion information, wherein the information integrates the recognition degree of various data types on each position and the corresponding effective feature judgment information.
Further, establishing a dynamic detection model includes:
collecting historical detection maintenance information, and preprocessing the historical detection maintenance information;
formulating a fault maintenance type according to the preprocessed historical detection maintenance information, wherein the fault maintenance type comprises corresponding fault parameter information;
and selecting a proper learning algorithm to learn the fault maintenance type, and establishing a dynamic detection model.
Specifically, historical detection maintenance information of the three-roller rotary rolling equipment is collected, including fault types, maintenance records, maintenance time and the like. And preprocessing such as data cleaning, missing value processing, abnormal value processing and the like is performed on the collected historical detection maintenance information, so that the accuracy and the integrity of the data are ensured. And detecting maintenance information according to the preprocessed history, analyzing fault types, and determining fault parameter information related to each fault type. For example, a certain fault type may be associated with parameters such as temperature, vibration, current, etc. And selecting a proper learning algorithm to learn the fault maintenance type and the corresponding fault parameter information thereof so as to establish a dynamic detection model, wherein common learning algorithms comprise decision trees, neural networks, support vector machines and the like. In the learning process, the preprocessed historical detection maintenance information is required to be used as a training set, and the relation between the fault maintenance type and the fault parameters is learned through a training algorithm, so that a dynamic detection model is established.
Further, inputting the dynamic feature fusion information into a dynamic detection model to obtain a fault detection result, including:
s610: matching the dynamic characteristic fusion information with the fault maintenance type in the dynamic detection model to obtain a dynamic matching result;
s620: setting a dynamic matching interval, judging whether a dynamic matching result is in the dynamic matching interval, and if so, obtaining a fault detection result according to the corresponding fault maintenance type; if not, the fault is marked as an in-doubt fault.
Specifically, the dynamic feature fusion information is used as input and matched with an established dynamic detection model. In the matching process, the dynamic feature fusion information is compared and analyzed with fault maintenance types in the model, the most relevant fault type is found out, and a dynamic matching result is obtained. A dynamic matching interval is set, and the interval can be subjected to statistical analysis based on historical data, such as calculating indexes of mean, variance, standard deviation and the like, and the dynamic matching interval is set based on the indexes. Checking whether the dynamic matching result is in a dynamic matching interval or not, and if the matching result is in the interval, namely, the set condition is met, obtaining a fault detection result according to the corresponding fault maintenance type; if the dynamic matching result is not in the dynamic matching interval, the dynamic matching result is marked as an in-doubt fault, and the in-doubt fault indicates that further confirmation and verification are needed.
Embodiment two:
based on the same inventive concept as the method for intelligently detecting faults of the three-roller rotary rolling equipment in the previous embodiment, the application also provides a system for intelligently detecting faults of the three-roller rotary rolling equipment, as shown in fig. 5, the system comprises:
static information acquisition module: collecting equipment parameters of three-roller rotary rolling equipment in a power-off state, and obtaining static multielement information;
static standard judging module: determining static standardized information of the three-roller rotary rolling equipment, judging whether the static multi-element information meets the static standardized information, if not, outputting the static multi-element information, and if so, executing the next step;
dynamic information acquisition module: collecting parameter information of the three-roller rotary rolling equipment in the working state, and obtaining dynamic multielement information;
and the characteristic information fusion module is used for: feature fusion is carried out on feature data in the dynamic multielement information to obtain dynamic feature fusion information;
and a fault detection module: establishing a dynamic detection model, inputting dynamic characteristic fusion information into the dynamic detection model, and obtaining a fault detection result; and overhauling the three-roller rotary rolling equipment through a fault detection result.
The adjusting system can effectively realize the method for intelligently detecting faults of the three-roller rotary rolling equipment, and has the technical effects as described in the embodiment, and the description is omitted here.
Further, the dynamic information acquisition module includes:
device cleaning unit: cleaning the three-roller rotary rolling equipment;
parameter acquisition unit: starting the three-roller rotary rolling equipment, and collecting multidimensional parameters of the three-roller rotary rolling equipment in a working state, wherein the multidimensional parameters can reflect the dynamic related state of the three-roller rotary rolling equipment;
extracting a characteristic unit: preprocessing the multidimensional parameter, and extracting dynamic characteristic information from the preprocessed multidimensional parameter;
an identification calculation unit: carrying out recognition calculation on the dynamic characteristic information and obtaining effective characteristic judgment information;
an acquisition output unit: and obtaining dynamic multielement information through the preprocessed multidimensional parameters and the effective characteristic judgment information.
Similarly, the above-mentioned optimization schemes of the system may also respectively correspond to the optimization effects corresponding to the methods in the first embodiment, which are not described herein again.
Although the application has been described in connection with specific features and embodiments thereof, it will be apparent that various modifications and combinations can be made without departing from the spirit and scope of the application. Accordingly, the specification and figures are merely exemplary illustrations of the present application as defined in the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (10)
1. The method for intelligently detecting faults of three-roller rotary rolling equipment is characterized by comprising the following steps of:
collecting equipment parameters of three-roller rotary rolling equipment in a power-off state, and obtaining static multielement information;
determining static standardized information of the three-roller rotary rolling equipment, judging whether the static multi-element information meets the static standardized information, if not, outputting the static multi-element information, and if so, executing the next step;
collecting parameter information of the three-roller rotary rolling equipment in the working state, and obtaining dynamic multielement information;
feature fusion is carried out on the feature data in the dynamic multivariate information to obtain dynamic feature fusion information;
establishing a dynamic detection model, and inputting the dynamic characteristic fusion information into the dynamic detection model to obtain a fault detection result;
and overhauling the three-roller rotary rolling equipment according to the fault detection result.
2. The method for intelligently detecting faults of three-roller rotary rolling equipment according to claim 1, wherein the steps of collecting equipment parameters of the three-roller rotary rolling equipment in a power-off state and obtaining static multivariate information comprise:
the data of each position of the processed three-roller rotary rolling equipment is acquired, and a static data subset is obtained by taking each position as a unit;
the lubrication detection is carried out on the three-roller rotary rolling equipment, and a lubrication detection result is obtained;
and obtaining the static multivariate information through the static data subsets of the positions and the lubrication detection results.
3. The method for intelligently detecting faults of the three-roller rotary rolling equipment according to claim 1, wherein the steps of collecting parameter information in the working state of the three-roller rotary rolling equipment and obtaining dynamic multielement information comprise:
starting the three-roller rotary rolling equipment, and collecting multi-dimensional parameters of the three-roller rotary rolling equipment in a working state, wherein the multi-dimensional parameters can reflect the dynamic related state of the three-roller rotary rolling equipment;
preprocessing the multi-dimensional parameters, and extracting dynamic characteristic information from the preprocessed multi-dimensional parameters;
carrying out recognition calculation on the dynamic characteristic information and obtaining effective characteristic judgment information;
and obtaining the dynamic multivariate information through the preprocessed multidimensional parameters and the effective characteristic judgment information.
4. The method for intelligently detecting faults of the three-roller rolling equipment according to claim 3, wherein the step of performing feature fusion on the feature data in the dynamic multivariate information to obtain dynamic feature fusion information comprises the following steps:
according to the effective characteristic judgment information corresponding to the data type in the multidimensional parameter, the identification degree of each effective characteristic judgment information to each position of the three-roller rolling equipment is obtained;
and carrying out feature fusion on the identification degree of each piece of corresponding effective feature judgment information, and carrying out correlation integration on different features to obtain the dynamic feature fusion information.
5. The method for intelligently detecting faults of three-roller rolling equipment according to claim 4, wherein the step of obtaining the identification degree of each effective characteristic judgment information on each position of the three-roller rolling equipment comprises the following steps:
the historical maintenance record of the three-roller rotary rolling equipment is called;
counting data characteristics in a plurality of data types according to the fault reasons in the historical maintenance records;
and the fault reasons correspond to data characteristics in a plurality of data types of the three-roller rotary rolling equipment and are mapped to the identification degree of the effective characteristic judgment information on each position of the three-roller rotary rolling equipment.
6. The method for intelligently detecting faults of three-roller rolling equipment according to claim 5, wherein the feature fusion is carried out on the identification degree of each piece of corresponding effective feature judgment information, and the correlation integration is carried out on different features to obtain the dynamic feature fusion information, and the method comprises the following steps:
the feature fusion adopts a weighted fusion algorithm, and the weighted fusion algorithm is as follows:
W＝a1*w1+a2*w2+……+ai*wi
wherein W is characteristic fusion information of one position of the three-roller rotary rolling equipment; a1 and a2 … … ai are the recognition degrees of the data types of various types to the position; w1 and w2 … … wi are respectively corresponding to the effective characteristic judgment information;
and collecting and integrating the characteristic fusion information of each position of the three-roller rotary rolling equipment to obtain the dynamic characteristic fusion information.
7. The method for intelligently detecting faults of the three-roller rotary rolling equipment according to claim 1, wherein the establishing a dynamic detection model comprises the following steps:
collecting historical detection maintenance information, and preprocessing the historical detection maintenance information;
formulating a fault maintenance type according to the preprocessed historical detection maintenance information, wherein the fault maintenance type comprises corresponding fault parameter information;
and selecting a proper learning algorithm to learn the fault maintenance type, and establishing a dynamic detection model.
8. The method for intelligently detecting faults of the three-roller rolling equipment according to claim 7, wherein the step of inputting the dynamic characteristic fusion information into the dynamic detection model to obtain fault detection results comprises the following steps:
the dynamic characteristic fusion information is matched with the fault maintenance type in the dynamic detection model, and a dynamic matching result is obtained;
setting a dynamic matching interval, judging whether the dynamic matching result is in the dynamic matching interval, and if so, acquiring the fault detection result according to the corresponding fault maintenance type; if not, the fault is marked as an in-doubt fault.
9. A system for intelligent fault detection of a three-roll rotary rolling device, the system comprising:
the static information acquisition module is used for acquiring equipment parameters of the three-roller rotary rolling equipment in a power-off state and acquiring static multielement information;
the static standard judging module is used for determining static standardized information of the three-roller rotary rolling equipment, judging whether the static multi-element information meets the static standardized information, outputting the static multi-element information if the static multi-element information does not meet the static standardized information, and executing the next step if the static multi-element information meets the static standardized information;
the dynamic information acquisition module is used for acquiring parameter information under the working state of the three-roller rotary rolling equipment and acquiring dynamic multielement information;
the feature information fusion module is used for carrying out feature fusion on the feature data in the dynamic multivariate information to obtain dynamic feature fusion information;
the fault detection module is used for establishing a dynamic detection model, inputting the dynamic characteristic fusion information into the dynamic detection model and obtaining a fault detection result; and overhauling the three-roller rotary rolling equipment according to the fault detection result.
10. The system for intelligently detecting faults of three-roller spinning equipment according to claim 9, wherein the dynamic information acquisition module comprises:
the parameter acquisition unit is used for starting the three-roller rotary rolling equipment and acquiring multi-dimensional parameters of the three-roller rotary rolling equipment in a working state, wherein the multi-dimensional parameters can reflect the dynamic relevant state of the three-roller rotary rolling equipment;
the feature extraction unit is used for preprocessing the multi-dimensional parameters and extracting dynamic feature information from the preprocessed multi-dimensional parameters;
the identification calculation unit is used for carrying out identification calculation on the dynamic characteristic information and obtaining effective characteristic judgment information;
and the output unit is used for obtaining the dynamic multivariate information through the preprocessed multidimensional parameter and the effective characteristic judgment information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310886772.9A CN117019889A (en) | 2023-07-18 | 2023-07-18 | Method and system for intelligently detecting faults of three-roller rotary rolling equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310886772.9A CN117019889A (en) | 2023-07-18 | 2023-07-18 | Method and system for intelligently detecting faults of three-roller rotary rolling equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117019889A true CN117019889A (en) | 2023-11-10 |
Family
ID=88643886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310886772.9A Pending CN117019889A (en) | 2023-07-18 | 2023-07-18 | Method and system for intelligently detecting faults of three-roller rotary rolling equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117019889A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5594713A (en) * | 1979-01-12 | 1980-07-18 | Kawasaki Steel Corp | Roll controller for steel shape mill |
| JPS6127110A (en) * | 1984-07-17 | 1986-02-06 | Kawasaki Steel Corp | Fault diagnosing method of hydraulic rolling reduction type rolling mill |
| JPH04361815A (en) * | 1991-06-06 | 1992-12-15 | Nippon Steel Corp | Abnormality diagnosing device for rolling-down device in rolling stages |
| JP2001265623A (en) * | 2000-03-17 | 2001-09-28 | Hitachi Ltd | Monitoring device |
| CN113468479A (en) * | 2021-06-16 | 2021-10-01 | 北京科技大学 | Data-driven cold continuous rolling industrial process monitoring and anomaly detection method |
| CN113762329A (en) * | 2021-07-06 | 2021-12-07 | 山东钢铁股份有限公司 | Method and system for constructing state prediction model of large rolling mill |
| CN113780852A (en) * | 2021-09-16 | 2021-12-10 | 东北大学 | Diagnosis method for quality defects in plate and strip rolling process |
| CN114492614A (en) * | 2022-01-20 | 2022-05-13 | 北京科技大学 | Method and device for classifying faults in hot rolling process of strip steel based on ensemble learning |
| CN114662856A (en) * | 2022-02-25 | 2022-06-24 | 山东钢铁股份有限公司 | Large-scale rolling mill state tracking method and system |
-
2023
- 2023-07-18 CN CN202310886772.9A patent/CN117019889A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5594713A (en) * | 1979-01-12 | 1980-07-18 | Kawasaki Steel Corp | Roll controller for steel shape mill |
| JPS6127110A (en) * | 1984-07-17 | 1986-02-06 | Kawasaki Steel Corp | Fault diagnosing method of hydraulic rolling reduction type rolling mill |
| JPH04361815A (en) * | 1991-06-06 | 1992-12-15 | Nippon Steel Corp | Abnormality diagnosing device for rolling-down device in rolling stages |
| JP2001265623A (en) * | 2000-03-17 | 2001-09-28 | Hitachi Ltd | Monitoring device |
| CN113468479A (en) * | 2021-06-16 | 2021-10-01 | 北京科技大学 | Data-driven cold continuous rolling industrial process monitoring and anomaly detection method |
| CN113762329A (en) * | 2021-07-06 | 2021-12-07 | 山东钢铁股份有限公司 | Method and system for constructing state prediction model of large rolling mill |
| CN113780852A (en) * | 2021-09-16 | 2021-12-10 | 东北大学 | Diagnosis method for quality defects in plate and strip rolling process |
| CN114492614A (en) * | 2022-01-20 | 2022-05-13 | 北京科技大学 | Method and device for classifying faults in hot rolling process of strip steel based on ensemble learning |
| CN114662856A (en) * | 2022-02-25 | 2022-06-24 | 山东钢铁股份有限公司 | Large-scale rolling mill state tracking method and system |
Non-Patent Citations (1)
| Title |
|---|
| 岳瑞华等: "《大坝风险综合评价理论、方法及应用》", 哈尔滨工业大学出版社, pages: 190 - 191 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112955839B (en) | Abnormality detection device, abnormality detection method, and program | |
| EP3764184B1 (en) | Abnormality determination assistance device | |
| CN113469060A (en) | Multi-sensor fusion convolution neural network aeroengine bearing fault diagnosis method | |
| CN103471841B (en) | A kind of rotating machinery vibrating failure diagnosis method | |
| CN113657221B (en) | Power plant equipment state monitoring method based on intelligent sensing technology | |
| CN111922095A (en) | Vibration diagnosis method for abnormal torsional vibration fault of roller of cold rolling mill | |
| CN112393906B (en) | Method for diagnosing, classifying and evaluating health of weak signal fault of bogie bearing of metro vehicle | |
| CN109187060B (en) | The detection of train speed sensor abnormal signal and axis locking method for diagnosing faults | |
| CN111611294A (en) | Star sensor data anomaly detection method | |
| CN112304611A (en) | Deep learning-based bearing fault diagnosis method | |
| WO2019043600A1 (en) | Remaining useful life estimator | |
| CN115758200A (en) | Vibration signal fault identification method and system based on similarity measurement | |
| CN116432071A (en) | Rolling bearing residual life prediction method | |
| CN111678699A (en) | Early fault monitoring and diagnosing method and system for rolling bearing | |
| CN115034094B (en) | Prediction method and system for operation state of metal processing machine tool | |
| Du et al. | Intelligent turning tool monitoring with neural network adaptive learning | |
| CN118134116B (en) | Cable and channel state monitoring and evaluating method based on big data analysis | |
| CN117350377A (en) | Knowledge graph driving-based equipment fault diagnosis method and device | |
| CN108393744A (en) | A kind of more sensor monitoring methods of cutting tool state | |
| CN118051822A (en) | Equipment running state abnormality detection method based on voiceprint recognition technology | |
| CN116910879A (en) | Cable force abnormality diagnosis method and device for bridge stay cable under random vehicle-mounted action | |
| CN117019889A (en) | Method and system for intelligently detecting faults of three-roller rotary rolling equipment | |
| CN108427375B (en) | Method for monitoring cutter state based on band-pass filtering processing multi-sensor | |
| CN113537156A (en) | Vibration data cleaning method based on interval standard deviation and spectrum analysis | |
| CN118518311B (en) | Impact detection equipment and method for hydrogen fuel cell |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |