CN117786373B - Equipment operation diagnosis system based on big data corrugated paper processing - Google Patents
Equipment operation diagnosis system based on big data corrugated paper processing Download PDFInfo
- Publication number
- CN117786373B CN117786373B CN202410218965.1A CN202410218965A CN117786373B CN 117786373 B CN117786373 B CN 117786373B CN 202410218965 A CN202410218965 A CN 202410218965A CN 117786373 B CN117786373 B CN 117786373B
- Authority
- CN
- China
- Prior art keywords
- data
- equipment
- node
- network
- speed
- 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.)
- Active
Links
- 238000012545 processing Methods 0.000 title claims abstract description 73
- 238000003745 diagnosis Methods 0.000 title claims abstract description 57
- 238000012423 maintenance Methods 0.000 claims abstract description 35
- 230000003993 interaction Effects 0.000 claims abstract description 30
- 238000005516 engineering process Methods 0.000 claims abstract description 23
- 230000015556 catabolic process Effects 0.000 claims abstract description 18
- 238000006731 degradation reaction Methods 0.000 claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 15
- 238000007405 data analysis Methods 0.000 claims abstract description 14
- 238000005457 optimization Methods 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims description 36
- 239000013598 vector Substances 0.000 claims description 36
- 230000006870 function Effects 0.000 claims description 22
- 230000004927 fusion Effects 0.000 claims description 20
- 230000005856 abnormality Effects 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 238000001514 detection method Methods 0.000 claims description 17
- 238000012544 monitoring process Methods 0.000 claims description 16
- 238000012706 support-vector machine Methods 0.000 claims description 16
- 238000003012 network analysis Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000013136 deep learning model Methods 0.000 claims description 12
- 238000005520 cutting process Methods 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 9
- 238000001035 drying Methods 0.000 claims description 9
- 230000007774 longterm Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 230000007613 environmental effect Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000010438 heat treatment Methods 0.000 claims description 6
- 230000015654 memory Effects 0.000 claims description 6
- 238000003825 pressing Methods 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000008447 perception Effects 0.000 claims description 4
- 239000002994 raw material Substances 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 230000006835 compression Effects 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000007500 overflow downdraw method Methods 0.000 claims description 3
- 238000003909 pattern recognition Methods 0.000 claims description 3
- 230000035945 sensitivity Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 3
- 230000001413 cellular effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 206010027175 memory impairment Diseases 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000013021 overheating Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 239000010687 lubricating oil Substances 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of equipment diagnosis, in particular to an equipment operation diagnosis system for corrugated paper processing based on big data, which comprises the following components: and a data acquisition module: collecting operation data of corrugated paper processing equipment in real time, including temperature, speed and pressure; and a data processing module: processing and analyzing the acquired data, and analyzing the running state of equipment; and a fault diagnosis module: based on the analysis result provided by the data processing module, combining with a big data analysis technology, identifying the potential failure and the reason of performance degradation of the equipment; a maintenance suggestion module: providing targeted maintenance and optimization suggestions according to the fault diagnosis result; user interaction interface: and displaying the running state of the equipment, fault diagnosis results and maintenance suggestions. The invention not only can capture the complex mode and the dependency relationship of the equipment operation in real time, but also can identify the abnormal state and the potential faults in time, thereby greatly improving the precision and the efficiency of fault diagnosis.
Description
Technical Field
The invention relates to the technical field of equipment diagnosis, in particular to an equipment operation diagnosis system for corrugated paper processing based on big data.
Background
In the corrugated paper processing industry, efficient and stable operation of equipment is a key for ensuring production efficiency and product quality. The traditional equipment maintenance method mainly depends on periodic inspection and experience judgment of operators, and the method is time-consuming and labor-consuming, and is difficult to accurately predict and timely discover potential faults of equipment, so that the equipment is stopped accidentally and the production efficiency is reduced. In addition, with the increase of corrugated paper demands and the complexity of production processes, the magnitude and complexity of equipment operation data are also greatly increased, and the traditional method is more worry.
While advances in information technology and automation technology have provided new means for equipment monitoring and fault diagnosis in recent years, such as collecting equipment operational data in real time by installing various sensors, and data monitoring and analysis using computer systems, these technologies have often focused on collecting and monitoring data, and there are still shortcomings in how to extract valuable information from a large amount of complex data, and accurately and rapidly diagnose equipment faults. Particularly, the lack of an effective data processing and analyzing model is difficult to deal with the high dimensionality and time series characteristics of equipment operation data, so that the accuracy and timeliness of fault diagnosis are insufficient.
In addition, the prior art is generally more general in terms of maintenance advice after fault diagnosis, and lacks pertinency and individuality, so that maintenance measures may not be accurate or effective enough, thereby affecting the efficiency and quality of equipment maintenance.
Disclosure of Invention
Based on the above object, the present invention provides an equipment operation diagnosis system for corrugated paper processing based on big data.
An equipment operation diagnosis system for corrugated paper processing based on big data, comprising:
And a data acquisition module: collecting operation data of corrugated paper processing equipment in real time, including temperature, speed and pressure;
and a data processing module: processing and analyzing the acquired data, and analyzing the running state of equipment;
And a fault diagnosis module: based on the analysis result provided by the data processing module, combining with a big data analysis technology, identifying the potential failure and the reason of performance degradation of the equipment;
A maintenance suggestion module: providing targeted maintenance and optimization suggestions according to the fault diagnosis result;
user interaction interface: and displaying the running state of the equipment, fault diagnosis results and maintenance suggestions.
Further, the data acquisition module specifically includes:
temperature sensor: the hot spot area comprises a drying section and a heating roller and is used for monitoring the temperature condition of the hot spot area in real time;
A speed sensor: the device is arranged on a conveying belt and a roller of a corrugated paper processing line and is used for measuring the movement speed of paper and equipment parts;
A pressure sensor: the pressure sensor is arranged on a pressure applying component in the corrugated paper processing process, the pressure component comprises a compression roller and a forming area, and the pressure sensor is designed for monitoring the pressure level in the corrugated paper processing process in real time so as to ensure the forming quality and the processing precision of the corrugated paper.
Further, the data processing module specifically includes:
data fusion technology: integrating data from different sensors using a data fusion technique to construct a panoramic view of the operation of the device, revealing complex relationships hidden behind the surface data by taking into account interactions between temperature, speed, pressure parameters;
The network analysis method comprises the following steps: mapping the operation state of the equipment into a dynamic network by using graph theory and a network analysis method, wherein nodes represent key components of the equipment, edges represent interactions or data flows among the key components, and influencing factors and potential fault points in the equipment are identified by analyzing the structure and dynamic changes of the network, and the key components comprise a hot roller, a corrugated roller, a pressing roller, a cutting machine, a drying section and a gumming machine;
Deep learning model: a deep learning model is adopted to directly learn complex representation from original data, and the complex representation is used for processing time series data and high-dimensional data, and deep modes and dependency relationships are learned from the data;
anomaly detection algorithm: and combining an abnormality detection algorithm to identify an abnormality mode in the equipment operation data in real time.
Further, the data fusion technique specifically includes:
the data from different sensors are standardized to eliminate the difference of dimension and magnitude, so that the data can be compared and fused on the same dimension, and the standardized formula is as follows: Wherein/> Is the original data,/>Is the average of the raw data,/>Is the standard deviation of the original data,/>Is the standardized data;
Weighted fusion: considering that the influence degree of different parameters (temperature, speed and pressure) on the running state of the equipment is different, adopting a weighted fusion method to integrate the different parameters, wherein the formula of the weighted fusion is as follows:
Wherein/> Is the characteristic value after fusion,/>、/>、/>Normalized values for temperature, velocity and pressure parameters, respectively,/>、/>、/>The weight coefficient of the corresponding parameter reflects the influence degree of each parameter on the running state of the equipment;
consideration of interaction terms: to reveal the interaction relationship between parameters, the interaction terms are introduced and the fusion model is updated as: Wherein/> 、/>And/>Is a weight coefficient of the interaction item, reflecting the degree of interaction between different parameters;
Fusing the characteristic values The method is used for monitoring the running state of equipment, diagnosing faults and evaluating performance.
Further, the network analysis method specifically includes:
dynamic network construction: determining key components of the device, each component acting as a node of the network, edges representing interactions or data flows between the nodes;
Network analysis: by utilizing the principle of graph theory, key influencing factors and potential fault points in equipment are identified by analyzing the structure and dynamic change of a network, wherein the key influencing factors and potential fault points comprise centrality and betweenness centrality, the centrality is used for measuring the connection number of one node and other nodes by the centrality, the nodes (components) with high centrality are key components for the equipment network, and the calculation formula of the centrality is as follows:
Wherein/> Is node/>Centrality of (v)/(v)Is node/>Degree of (i.e. directly connected to/>)Node number)/>Is the total number of nodes in the network;
the medium centrality is used for measuring the occurrence frequency of a node on all shortest paths in the network, and for the equipment network, the node with high medium centrality is a key path for fault propagation, and the calculation formula of the medium centrality is as follows: Wherein/> Is node/>Medium centrality of/>Is node/>To node/>Total number of shortest paths between/(Is the path through node/>Is a number of (3).
Further, the deep learning model adopts an improved long-short-term memory network LSTM model, an environment sensing door is introduced, the sensitivity of the model to external environment changes and equipment operation condition changes is enhanced, and the improved long-short-term memory network LSTM model specifically comprises:
Forgetting the door: control of last state information/> Is the forgetfulness of (a);
An input door: control the current input/> Is a reception degree of (2);
Candidate cell status: Generating candidate cell states, including new information;
Environmental perception gate: according to the environment and operating conditions/> Adjusting the flow of information;
Cell status update: Updating the current cell state, and integrating the influence of a forgetting gate, an input gate and an environment sensing gate;
Output door: Control from cellular State/> To final output/>Is a stream of information;
Final output: determining the output at the moment according to the output gate and the current cell state;
wherein;
at time/> Inputting a data vector;
: the output vector of the last time step represents the information of the last state;
: the cell state vector of the last time step stores long-term information;
at time/> External environment and operating condition vectors of (a);
forget gate, input gate, environment aware gate and output gate at time/> An activation value of (2);
time/> Is a candidate cell state of (a);
Time cell state vector, and storing updated long-term information;
time/> Is an output vector representing information of the current state;
: the weight matrixes are respectively a forgetting gate, an input gate, a candidate cell state, an environment sensing gate and an output gate;
: bias vectors of the forgetting gate, the input gate, the candidate cell state, the environment sensing gate and the output gate respectively;
a Sigmoid function for the activation of the gate controller, mapping the input between 0 and 1;
tanh: hyperbolic tangent function for activation and output mapping the input between 1 and 1.
Further, the anomaly detection algorithm adopts an isolated forest algorithm to perform anomaly detection, and identifies an anomaly mode in the equipment operation data in real time, and specifically comprises the following steps:
Constructing an isolated tree: for each sample in the data set, randomly selecting a feature, then randomly selecting a cutting point between the maximum value and the minimum value of the feature, recursively repeating until each sample is isolated to form an isolated tree, and repeatedly constructing a plurality of isolated trees to form an isolated forest;
Calculating an anomaly score: the path length of the sample in the isolated tree is used for calculating an anomaly score, and the shorter the path is, the closer the sample is to the anomaly value;
Abnormality determination: and determining that the sample is abnormal if the score of the sample exceeds a predetermined threshold value according to the calculated abnormal score.
Further, the calculation formula of the anomaly score is as follows:
Wherein/> Is a sample/>Abnormality score of/>Is a sample/>Average path length over all orphaned trees,/>Is the number of samples,/>Is the number of samples in the dataset/>Average unsuccessfully isolated path length.
Further, the fault diagnosis module specifically includes:
Receiving an output of a data processing module: the fault diagnosis module receives analysis results from the data processing module, including a network analysis method, anomaly detection and the output of a deep learning model, reflects the running states and behavior modes of key components of the equipment, and combines the analysis results from different sources into a comprehensive feature set;
Applying big data analysis technology: analyzing the integrated comprehensive feature set by utilizing a big data analysis technology to identify and classify the normal running state, potential failure mode and signs of performance degradation of the equipment;
fault mode identification: based on the historical fault data and the case library, the fault diagnosis module matches the current abnormal mode with the known fault mode to determine potential fault types and reasons;
performance degradation analysis: for non-abrupt performance degradation problems, the fault diagnosis module will analyze the long-term trends and pattern changes to identify potential factors that lead to performance degradation;
Outputting a diagnosis result: the fault diagnostic module provides a diagnostic report.
Further, the big data analysis technology is based on a Support Vector Machine (SVM), which specifically comprises:
the comprehensive feature sets are fused into a comprehensive feature vector, and the comprehensive feature vector is subjected to standardization or normalization processing to ensure that all features have similar scales;
training an SVM model: the kernel function selects the RBF kernel, denoted as: Wherein/> And/>Is two eigenvectors,/>The method is characterized in that the parameters of RBF cores are used for controlling the distribution of the mapped feature space, the SVM is set to find out a hyperplane, the interval between different types of data is maximized, and based on the two classification problems, the calculation formula is as follows:
；
；
Wherein, Is the normal vector of the hyperplane,/>Is an offset term,/>Is a regularization parameter,/>Is a variable of the relaxation,Is a function of mapping the input vector into a high-dimensional space;
Fault pattern recognition and classification: classifying the new comprehensive feature set by using a trained SVM model to identify the running state of the equipment, wherein a decision function of the SVM model is used for classifying decisions, and the expression is as follows: wherein/> Is Lagrangian multiplier,/>Is a class label of training samples,/>Is the result of kernel function computation,/>Is a sign function for outputting the classification result.
The invention has the beneficial effects that:
According to the invention, the data processing module adopts advanced technologies such as an improved long and short term memory network (LSTM) and an isolated forest algorithm, and can effectively process and analyze a large amount of operation data from corrugated paper processing equipment, and through the technologies, the module can not only capture the complex mode and the dependency relationship of the equipment operation in real time, but also can timely identify abnormal states and potential faults, and greatly improve the accuracy and efficiency of fault diagnosis.
According to the invention, the fault diagnosis module is combined with the big data analysis technology, and potential faults and performance degradation reasons of the equipment can be accurately identified based on the deep analysis result provided by the data processing module. The deep fault diagnosis capability enables a maintenance team to formulate maintenance and optimization strategies in a targeted manner, avoids invalid maintenance and excessive maintenance, and improves the pertinence and effect of maintenance work.
According to the invention, through real-time and accurate fault diagnosis and targeted maintenance suggestion, the corrugated paper processing enterprise can be helped to solve the equipment problem in time, the unexpected downtime of the equipment is reduced, the service life of the equipment is prolonged, and the overall production efficiency and the product quality are improved. In addition, the predictive maintenance function of the system can also help enterprises optimize resource allocation and maintenance plans, and further reduce operation cost.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only of the invention and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of a system functional module according to an embodiment of the present invention.
Detailed Description
The present invention will be further described in detail with reference to specific embodiments in order to make the objects, technical solutions and advantages of the present invention more apparent.
It is to be noted that unless otherwise defined, technical or scientific terms used herein should be taken in a general sense as understood by one of ordinary skill in the art to which the present invention belongs. The terms "first," "second," and the like, as used herein, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
As shown in fig. 1, an equipment operation diagnosis system for corrugated paper processing based on big data, comprising:
And a data acquisition module: collecting operation data of corrugated paper processing equipment in real time, including temperature, speed and pressure;
and a data processing module: processing and analyzing the acquired data, and analyzing the running state of equipment;
And a fault diagnosis module: based on the analysis result provided by the data processing module, combining with a big data analysis technology, identifying the potential failure and the reason of performance degradation of the equipment;
A maintenance suggestion module: providing targeted maintenance and optimization suggestions according to the fault diagnosis result;
user interaction interface: and displaying the running state of the equipment, fault diagnosis results and maintenance suggestions.
The data acquisition module specifically comprises:
temperature sensor: the hot spot area comprises a drying section and a heating roller and is used for monitoring the temperature condition of the hot spot area in real time;
A speed sensor: the device is arranged on a conveying belt and a roller of a corrugated paper processing line and is used for measuring the movement speed of paper and equipment parts;
A pressure sensor: the pressure sensor is arranged on a pressure applying component in the corrugated paper processing process, the pressure component comprises a compression roller and a forming area, and the pressure sensor is designed for monitoring the pressure level in the corrugated paper processing process in real time so as to ensure the forming quality and the processing precision of the corrugated paper.
The data acquisition module also comprises a data summarization unit which is responsible for collecting the data of each sensor and transmitting it to the data processing module in a standardized format. The data summarizing unit adopts a high-speed data communication technology to ensure that the data collected from each sensor can be transmitted and processed without delay, thereby realizing the real-time monitoring of the running state of the corrugated paper processing equipment.
The data processing module specifically comprises:
data fusion technology: integrating data from different sensors using a data fusion technique to construct a panoramic view of the operation of the device, revealing complex relationships hidden behind the surface data by taking into account interactions between temperature, speed, pressure parameters;
The network analysis method comprises the following steps: mapping the operation state of the equipment into a dynamic network by using graph theory and a network analysis method, wherein nodes represent key components of the equipment, edges represent interactions or data flows among the key components, and influencing factors and potential fault points in the equipment are identified by analyzing the structure and dynamic changes of the network, and the key components comprise a hot roller, a corrugated roller, a pressing roller, a cutting machine, a drying section and a gumming machine;
Deep learning model: a deep learning model is adopted to directly learn complex representation from original data, and the complex representation is used for processing time series data and high-dimensional data, and deep modes and dependency relationships are learned from the data;
anomaly detection algorithm: and combining an abnormality detection algorithm to identify an abnormality mode in the equipment operation data in real time.
The data fusion technology specifically comprises the following steps:
the data from different sensors are standardized to eliminate the difference of dimension and magnitude, so that the data can be compared and fused on the same dimension, and the standardized formula is as follows: Wherein/> Is the original data,/>Is the average of the raw data,/>Is the standard deviation of the original data,/>Is the standardized data;
Weighted fusion: considering that the influence degree of different parameters (temperature, speed and pressure) on the running state of the equipment is different, adopting a weighted fusion method to integrate the different parameters, wherein the formula of the weighted fusion is as follows:
Wherein/> Is the characteristic value after fusion,/>、/>、/>Normalized values for temperature, velocity and pressure parameters, respectively,/>、/>、/>The weight coefficient of the corresponding parameter reflects the influence degree of each parameter on the running state of the equipment;
consideration of interaction terms: to reveal the interaction relationship between parameters, the interaction terms are introduced and the fusion model is updated as: Wherein/> 、/>And/>Is a weight coefficient of the interaction item, reflecting the degree of interaction between different parameters;
Fusing the characteristic values The method is used for monitoring the running state of equipment, diagnosing faults and evaluating performance.
Weight coefficient、/>、/>、/>、/>/>Training and optimizing through historical data to ensure the fused characteristic value/>Reflecting the actual operating state of the device to the greatest extent.
The network analysis method specifically comprises the following steps:
dynamic network construction: determining key components of the device, each component acting as a node of the network, e.g. node a representing a hot roll and node B representing a cutter; edges represent interactions or data flows between nodes, and if the output of component A is the input of component B, there is an edge from node A to node B. Edges may be directional (representing the direction of data or physical flow) or undirected (representing bi-directional interactions);
Network analysis: by utilizing the principle of graph theory, key influencing factors and potential fault points in equipment are identified by analyzing the structure and dynamic change of a network, wherein the key influencing factors and potential fault points comprise centrality and betweenness centrality, the centrality is used for measuring the connection number of one node and other nodes by the centrality, and for the equipment network, the nodes (components) with high centrality are key components because of interaction with a plurality of other components, and the calculation formula of the centrality is as follows:
Wherein/> Is node/>Centrality of (v)/(v)Is node/>Degree of (i.e. directly connected to/>)Node number)/>Is the total number of nodes in the network;
The betweenness is used for measuring the occurrence frequency of a node on all shortest paths in the network, and for the equipment network, the node with high betweenness is a key path for fault propagation, and the calculation formula of the betweenness is as follows: Wherein/> Is node/>Medium centrality of/>Is node/>To node/>Total number of shortest paths between/(Is the path through node/>Is a number of (3).
The deep learning model adopts an improved long-short-term memory network LSTM model, an environment sensing door is introduced, the sensitivity of the model to external environment changes and equipment operation condition changes is enhanced, and the improved long-short-term memory network LSTM model specifically comprises:
Forgetting the door: control of last state information/> Is the forgetfulness of (a);
An input door: control the current input/> Is a reception degree of (2);
Candidate cell status: Generating candidate cell states, including new information;
Environmental perception gate: according to the environment and operating conditions/> Adjusting the flow of information;
Cell status update: Updating the current cell state, and integrating the influence of a forgetting gate, an input gate and an environment sensing gate;
Output door: Control from cellular State/> To final output/>Is a stream of information;
Final output: determining the output at the moment according to the output gate and the current cell state;
wherein;
at time/> Inputting data vectors, including equipment operation parameters such as temperature, speed, pressure and the like;
: the output vector of the last time step represents the information of the last state;
: the cell state vector of the last time step stores long-term information;
at time/> External environment and operation condition vectors such as raw material properties, environment temperature and humidity, etc.;
forget gate, input gate, environment aware gate and output gate at time/> An activation value of (2);
time/> Is a candidate cell state of (a);
Time cell state vector, and storing updated long-term information;
time/> Is an output vector representing information of the current state;
: the weight matrixes are respectively a forgetting gate, an input gate, a candidate cell state, an environment sensing gate and an output gate;
: bias vectors of the forgetting gate, the input gate, the candidate cell state, the environment sensing gate and the output gate respectively;
a Sigmoid function for the activation of the gate controller, mapping the input between 0 and 1;
tanh: hyperbolic tangent function for activation and output mapping the input between 1 and 1.
The LSTM model can dynamically adjust the flow of information according to the current environmental conditions and operation parameters through the introduction of the environment sensing door, the adaptability of the model to environmental changes is enhanced, for example, in the corrugated paper processing process, the changes of the humidity, the temperature conditions or the machine operation speed of raw materials can influence the product quality and the equipment running state, and the EAG can capture the changes and correspondingly adjust the behavior of the model.
Cell status update improvement: by adding the output of the context aware gate to the update of the cell state, the model is able to take into account the specific effects of the external environment and operating conditions at each time step, making the state update more accurate and targeted. The mechanism is particularly suitable for handling complex situations in the operation of corrugated paper processing equipment, such as differences between raw material batches, environmental fluctuations caused by seasonal changes, and the like.
Through the improvement, the LSTM network can more accurately capture and predict the running states of the corrugated paper processing equipment under different environments and operating conditions, and support is provided for equipment fault prediction, maintenance planning and production optimization. The innovative LSTM model not only can process time series data and high-dimensional data, but also can adapt to complex production environments, and improves equipment management and operation efficiency.
The anomaly detection algorithm adopts an isolated forest algorithm to detect anomalies and identify anomaly modes in equipment operation data in real time, and the isolated forest algorithm randomly selects a feature and a random score value of the feature to isolate data points, so that the anomaly points are easy to isolate, and the method has a short path and specifically comprises the following steps:
Constructing an isolated tree: for each sample in the data set, randomly selecting a feature, then randomly selecting a cutting point between the maximum value and the minimum value of the feature, recursively repeating until each sample is isolated to form an isolated tree, and repeatedly constructing a plurality of isolated trees to form an isolated forest;
Calculating an anomaly score: the path length of the sample in the isolated tree is used for calculating an anomaly score, and the shorter the path is, the closer the sample is to the anomaly value;
Abnormality determination: and determining that the sample is abnormal if the score of the sample exceeds a predetermined threshold value according to the calculated abnormal score.
When an isolated forest algorithm is applied to the operation data of the corrugated paper processing equipment, monitoring and identifying abnormal modes in the equipment state data in real time, randomly selecting a sub-sample from the operation data set of the corrugated paper processing equipment, and constructing an isolated tree on the sub-sample:
data preprocessing: raw data (temperature, speed, pressure, etc.) collected from the corrugated paper processing equipment is first pre-processed, including cleaning, normalization, to accommodate the input requirements of the isolated forest algorithm.
Dynamic subsample selection: to accommodate the real-time and dynamic nature of the equipment operational data, subsamples may be randomly selected from the most recent operational data periodically or based on specific events to build an isolated forest.
Real-time anomaly detection: by calculating the abnormality score for the real-time data of the equipment, the abnormality mode in the running state of the equipment can be rapidly identified, and when the abnormality score exceeds a preset threshold value, an alarm is triggered to prompt maintenance personnel to check and intervene.
Continuous learning and adaptation: over time, the orphan forest model can be updated continuously according to the newly collected data, and the detection capability of the newly-appearing abnormal mode is improved.
The calculation formula of the anomaly score is as follows:
Wherein/> Is a sample/>Abnormality score of/>Is a sample/>Average path length over all orphaned trees,/>Is the number of samples,/>Is the number of samples in the dataset/>Average unsuccessfully isolated path length.
The fault diagnosis module specifically comprises:
Receiving an output of a data processing module: the fault diagnosis module receives analysis results from the data processing module, including a network analysis method, anomaly detection and the output of a deep learning model, reflects the running states and behavior modes of key components of the equipment, and combines the analysis results from different sources into a comprehensive feature set;
Applying big data analysis technology: analyzing the integrated comprehensive feature set by utilizing a big data analysis technology to identify and classify the normal running state, potential failure mode and signs of performance degradation of the equipment;
fault mode identification: based on the historical fault data and the case library, the fault diagnosis module matches the current abnormal mode with the known fault mode to determine potential fault types and reasons;
performance degradation analysis: for non-abrupt performance degradation problems, the fault diagnosis module will analyze the long-term trends and pattern changes to identify potential factors that lead to performance degradation;
Outputting a diagnosis result: the fault diagnostic module provides a diagnostic report.
Assuming that the data processing module identifies an anomaly pattern by an isolated forest algorithm, the pattern indicates that an unusual pressure fluctuation occurs in the pressure roller portion of the corrugated paper processing apparatus, the fault diagnosis module will receive the anomaly detection result and analyze in combination with historical operating data, maintenance records and similar fault cases of the pressure roller. By applying a classification algorithm, the fault diagnosis module may recognize that such pressure fluctuations have a pattern similar to some past fault events (e.g., roll surface wear, bearing damage, etc.).
Further, the fault diagnostic module will analyze the operational data of the relevant components, such as temperature, speed and status of other components interacting with the pressure roller, to determine the specific cause of the fault, for example, if an abnormally elevated temperature is simultaneously found, possibly indicating a pressure roller fault caused by overheating of the bearing.
Eventually, the fault diagnostic module will output a diagnostic report indicating that the abnormal pressure fluctuations of the pressure roll may be caused by overheating of the bearing, suggesting that bearing inspection and necessary maintenance be performed.
The big data analysis technology is based on a Support Vector Machine (SVM), which specifically comprises:
the comprehensive feature sets are fused into a comprehensive feature vector, and the comprehensive feature vector is subjected to standardization or normalization processing to ensure that all features have similar scales;
training an SVM model: the kernel function selects the RBF kernel, denoted as: Wherein/> And/>Is two eigenvectors,/>The method is characterized in that the parameters of RBF cores are used for controlling the distribution of the mapped feature space, the SVM is set to find out a hyperplane, the interval between different types of data is maximized, and based on the two classification problems, the calculation formula is as follows:
；
；/>
Wherein, Is the normal vector of the hyperplane,/>Is an offset term,/>Is a regularization parameter,/>Is a variable of the relaxation,Is a function of mapping the input vector into a high-dimensional space;
Fault pattern recognition and classification: classifying the new comprehensive feature set by using a trained SVM model to identify the running state of the equipment, wherein a decision function of the SVM model is used for classifying decisions, and the expression is as follows: wherein/> Is Lagrangian multiplier,/>Is a class label of training samples,/>Is the result of kernel function computation,/>Is a sign function for outputting the classification result.
In the context of fault diagnosis of corrugated paper processing equipment, the SVM model may classify the operational state of the equipment as "normal", "latent fault" or "performance degradation" based on the fused feature set. For example, if the SVM model identifies that a state corresponding to a feature vector is "latent fault," the abnormal features in the feature vector may be further analyzed, and the specific type and possible cause of the fault may be determined in combination with the operational data and maintenance records of the device. The method not only improves the accuracy of fault diagnosis, but also can early warn before the fault occurs, and provides powerful support for equipment maintenance and performance optimization.
The fault diagnosis result includes:
Pressure roller anomaly: if the diagnostic result indicates that there is abnormal pressure fluctuation in the pressure roller, it may indicate a loss of bearing or a failure of bearing.
Temperature fluctuation: if the diagnostic results indicate that there is an unstable fluctuation in the temperature of the heating section, it may be a heating system failure or a temperature control system failure.
Speed inconsistency: if the diagnostic result indicates that the corrugator machine is not operating at the same speed, it may be a driveline problem or motor failure.
The cutting accuracy is reduced: if the diagnostic results indicate a reduced accuracy of the cutter, it may be a problem with cutter wear or cutter calibration.
The maintenance suggestion module includes:
For pressure roller anomalies:
Proposal: damaged bearings are inspected and replaced, and bearing lubrication is performed periodically to reduce wear.
Optimizing: higher performance bearings are installed or lubricating oils more suitable for high load operating conditions are used.
For temperature fluctuations:
Proposal: the electrical connections and components of the heating system are checked to ensure that the temperature sensor is properly calibrated.
Optimizing: the temperature control system is upgraded, and a more advanced PID controller or an adaptive control algorithm is introduced.
For speed inconsistencies:
proposal: the belt or chain is inspected for tension and wear and the damaged drive assembly is replaced.
Optimizing: and installing a speed monitoring system, and adjusting the output of the motor in real time to keep a stable running speed.
For a decrease in cutting accuracy:
proposal: the worn tool is replaced and the cutter is periodically calibrated to ensure accuracy.
Optimizing: an automatic tool wear monitoring and calibration system is introduced to reduce human intervention.
The maintenance recommendation module should include a knowledge base containing various failure modes and their corresponding maintenance measures and optimization recommendations, and when the failure diagnosis module identifies a particular failure mode, the maintenance recommendation module will retrieve the relevant recommendation from the knowledge base and provide it to the operator or maintenance team.
Those of ordinary skill in the art will appreciate that: the discussion of any of the embodiments above is merely exemplary and is not intended to suggest that the scope of the invention is limited to these examples; the technical features of the above embodiments or in the different embodiments may also be combined within the idea of the invention, the steps may be implemented in any order and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.
The present invention is intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent replacement, improvement, etc. of the present invention should be included in the scope of the present invention.
Claims (6)
1. An equipment operation diagnosis system for corrugated paper processing based on big data, comprising:
And a data acquisition module: collecting operation data of corrugated paper processing equipment in real time, including temperature, speed and pressure;
and a data processing module: processing and analyzing the collected data, and analyzing the running state of equipment, wherein the data processing module specifically comprises:
data fusion technology: integrating data from different sensors using a data fusion technique to construct a panoramic view of the operation of the device, revealing complex relationships hidden behind surface data by considering interactions between temperature, speed, pressure parameters, the data fusion technique specifically comprising:
the data from different sensors are standardized to eliminate the difference of dimension and magnitude, so that the data can be compared and fused on the same dimension, and the standardized formula is as follows: where x is the raw data, μ is the average of the raw data, σ is the standard deviation of the raw data, and z is the normalized data;
weighted fusion: considering that the influence degrees of different parameters on the running state of the equipment are different, adopting a weighted fusion method to integrate the different parameters, wherein the weighted fusion formula is as follows:
F=w 1·Z Temperature (temperature) +w2·Z Speed of speed +w3·Z Pressure of , wherein F is a fused characteristic value, Z Temperature (temperature) 、Z Speed of speed 、Z Pressure of is a standardized value of temperature, speed and pressure parameters, and w 1、w2、w3 is a weight coefficient of the corresponding parameters, reflecting the influence degree of each parameter on the running state of the equipment;
consideration of interaction terms: to reveal the interaction relationship between parameters, the interaction terms are introduced and the fusion model is updated as:
F＝w1·Z Temperature (temperature) +w2·Z Speed of speed +w3·Z Pressure of +w4·Z Temperature (temperature) ·Z Speed of speed +w5·Z Temperature (temperature) ·Z Pressure of +w6·Z Speed of speed ·Z Pressure of , Wherein w 4、w5 and w 6 are weight coefficients of the interaction terms reflecting the degree of interaction between different parameters;
the fused characteristic value F is used for monitoring the running state of the equipment, diagnosing faults and evaluating performance;
The network analysis method comprises the following steps: mapping the operation state of the equipment into a dynamic network by using graph theory and a network analysis method, wherein nodes represent key components of the equipment, edges represent interactions or data flows among the key components, influencing factors and potential fault points in the equipment are identified by analyzing the structure and dynamic changes of the network, and the key components comprise a hot roller, a corrugated roller, a pressing roller, a cutting machine, a drying section and a gumming machine, and the network analysis method specifically comprises the following steps:
Dynamic network construction: the key components of the equipment are a hot roller, a corrugated roller, a pressing roller, a cutting machine, a drying section and a gumming machine, wherein the hot roller, the corrugated roller, the pressing roller, the cutting machine, the drying section and the gumming machine are respectively used as a node of a network, and edges represent interaction or data flow among the nodes;
Network analysis: by utilizing the principle of graph theory, key influencing factors and potential fault points in equipment are identified by analyzing the structure and dynamic change of a network, wherein the key influencing factors comprise degree centrality and betweenness centrality, the degree centrality is used for measuring the connection number of one node and other nodes, the node with high centrality is a key component for the equipment network, and a calculation formula of the degree centrality is as follows:
Wherein C D (v) is the degree centrality of the node v, deg (v) is the degree of the node v, namely the number of nodes directly connected to v, the node v comprises a hot roll, a corrugating roll, a pressing roll, a cutting machine, a drying section and a gumming machine, n is the total number of nodes in the network, namely n=6, and the degree centrality is obtained by inputting deg (v) and n to output C D (v), namely the ratio of the degree of the node v to the maximum degree (n-1) in the network, reflecting the connection degree of the node v in the network;
the medium centrality is used for measuring the occurrence frequency of a node on all shortest paths in the network, and for the equipment network, the node with high medium centrality is a key path for fault propagation, and the calculation formula of the medium centrality is as follows: Where C B (v) is the median centrality of node v, σ st is the total number of shortest paths between the start node s and the target node t, σ st (v) is the number of paths through node v in the shortest path between the start node s and the target node t, and the median centrality is determined by inputting σ st (v) to output a value of C B (v), with higher values indicating greater influence of node v in the network because more shortest paths depend on the node;
Deep learning model: a deep learning model is adopted to directly learn complex representation from original data, and the complex representation is used for processing time series data and high-dimensional data, and deep modes and dependency relationships are learned from the data;
Anomaly detection algorithm: combining an abnormality detection algorithm, and identifying an abnormality mode in the equipment operation data in real time;
and a fault diagnosis module: based on the analysis result provided by the data processing module, combining with big data analysis technology, identifying the potential faults and reasons for performance degradation of the equipment, wherein the fault diagnosis module specifically comprises:
Receiving an output of a data processing module: the fault diagnosis module receives analysis results from the data processing module, including a network analysis method, anomaly detection and the output of a deep learning model, reflects the running states and behavior modes of key components of the equipment, and combines the analysis results from different sources into a comprehensive feature set;
Applying big data analysis technology: analyzing the integrated comprehensive feature set by utilizing a big data analysis technology to identify and classify the normal running state, potential failure mode and signs of performance degradation of the equipment;
fault mode identification: based on the historical fault data and the case library, the fault diagnosis module matches the current abnormal mode with the known fault mode to determine potential fault types and reasons;
performance degradation analysis: for non-abrupt performance degradation problems, the fault diagnosis module will analyze the long-term trends and pattern changes to identify potential factors that lead to performance degradation;
outputting a diagnosis result: the fault diagnosis module provides a diagnosis report;
A maintenance suggestion module: providing targeted maintenance and optimization suggestions according to the fault diagnosis result;
user interaction interface: and displaying the running state of the equipment, fault diagnosis results and maintenance suggestions.
2. The equipment operation diagnosis system for corrugated paper processing based on big data according to claim 1, wherein the data acquisition module specifically comprises:
temperature sensor: the hot spot area comprises a drying section and a heating roller and is used for monitoring the temperature condition of the hot spot area in real time;
A speed sensor: the device is arranged on a conveying belt and a roller of a corrugated paper processing line and is used for measuring the movement speed of paper and equipment parts;
A pressure sensor: the pressure sensor is arranged on a pressure applying component in the corrugated paper processing process, the pressure component comprises a compression roller and a forming area, and the pressure sensor is designed for monitoring the pressure level in the corrugated paper processing process in real time so as to ensure the forming quality and the processing precision of the corrugated paper.
3. The equipment operation diagnosis system based on big data corrugated paper processing according to claim 2, wherein the deep learning model adopts an improved long-short-term memory network LSTM model, an environment sensing door is introduced, sensitivity of the model to external environment changes and equipment operation condition changes is enhanced, and the improved long-short-term memory network LSTM model specifically comprises:
Forgetting the door: f t＝σ(Wf·[ht-1,xt]+bf) controlling the forgetting degree of the previous state information C t-1;
An input door: i t＝σ(Wi·[ht-1,xt]+bi), control the current input Is a reception degree of (2);
Candidate cell status: generating candidate cell states, including new information;
Environmental perception gate: g t＝σ(Wg·[ht-1,xt,et]+bg) to adjust the flow of information according to environmental and operating conditions e t;
Cell status update: Updating the current cell state, and integrating the influence of a forgetting gate, an input gate and an environment sensing gate;
Output door: o t＝σ(Wo·[ht-1,xt]+bo), controlling the flow of information from cell state C t to final output h t;
final output: h t＝ot*tanh(Ct) determining the output at this time according to the output gate and the current cell state;
wherein;
x t: inputting data vectors, namely temperature, speed and pressure parameters at time t;
h t-1: the output vector of the last time step represents the information of the last state;
c t-1: the cell state vector of the last time step stores long-term information;
e t: the external environment and operation condition vector at time t comprises raw material properties and environment temperature and humidity;
f t,it,gt,ot: forget the activation value of door, input door, environment perception door and output door at time t;
Candidate cell status at time t;
C t: time-witnessing the cell state vector, and storing updated long-term information;
h t: the output vector of time t represents the information of the current state and is used for evaluating the current running state of the equipment and judging whether the equipment runs normally or has performance deviation;
w f,Wi,WC,Wg,Wo: the weight matrixes are respectively a forgetting gate, an input gate, a candidate cell state, an environment sensing gate and an output gate;
b f,bi,bC,bg,bo: bias vectors of the forgetting gate, the input gate, the candidate cell state, the environment sensing gate and the output gate respectively;
sigma: a Sigmoid function for activation of the gate controller, mapping the input between 0 and 1;
tan h: the hyperbolic tangent function, used for activation and output, maps the input between 1 and 1.
4. The equipment operation diagnosis system for corrugated paper processing based on big data according to claim 3, wherein the abnormality detection algorithm adopts an isolated forest algorithm for abnormality detection, and identifies an abnormality mode in the equipment operation data in real time, specifically comprising:
Constructing an isolated tree: for each sample in the data set containing the temperature, speed and pressure characteristics, randomly selecting and inputting a characteristic, then randomly selecting a cutting point between the maximum value and the minimum value of the characteristic, recursively repeating until each sample is isolated to form an isolated tree, and repeatedly constructing a plurality of isolated trees to form an isolated forest;
Calculating an anomaly score: the path length of the sample in the isolated tree is used for calculating an anomaly score, and the shorter the path is, the closer the sample is to the anomaly value;
Abnormality determination: if the score of the sample exceeds a predetermined threshold, the sample is determined to be abnormal based on the output abnormality score.
5. The equipment operation diagnosis system for corrugated paper processing based on big data according to claim 4, wherein the calculation formula of the anomaly score is:
Where s (x, n) is the anomaly score for sample x, E (h (x)) is the average path length of sample x over all orphaned trees, n is the number of samples, and c (n) is the average unsuccessfully orphaned path length for a number of samples in the dataset of n.
6. The system for diagnosing equipment operation for corrugated paper processing based on big data as recited in claim 5, wherein the big data analysis technique is based on a support vector machine SVM, and the support vector machine SVM specifically comprises:
the comprehensive feature sets are fused into a comprehensive feature vector, and the comprehensive feature vector is subjected to standardization or normalization processing to ensure that all features have similar scales;
Training an SVM model: the kernel function selects RBF kernels, which are expressed as K (x i,xj)＝exp(-γ||xi-xj||2), wherein x i and x j are two feature vectors, gamma is a parameter of the RBF kernels, the distribution of the mapped feature space is controlled, the goal of the SVM is to find a hyperplane, the interval between different types of data is maximized, and a calculation formula is as follows based on the two classification problem:
Where w is the normal vector to the hyperplane, b is the bias term, C is the regularization parameter, ζ i is the relaxation variable, Is a function of mapping the input vector into a high-dimensional space;
Fault pattern recognition and classification: classifying the new comprehensive feature set by using a trained SVM model to identify the running state of the equipment, wherein a decision function of the SVM model is used for classifying decisions, and the expression is as follows: Where α i is the Lagrangian multiplier, y i is the class label of the training sample, K (x i, x) is the result of kernel function calculation, sgn (·) is the sign function, and is used to output the classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410218965.1A CN117786373B (en) | 2024-02-28 | 2024-02-28 | Equipment operation diagnosis system based on big data corrugated paper processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410218965.1A CN117786373B (en) | 2024-02-28 | 2024-02-28 | Equipment operation diagnosis system based on big data corrugated paper processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117786373A CN117786373A (en) | 2024-03-29 |
| CN117786373B true CN117786373B (en) | 2024-05-03 |
Family
ID=90385702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410218965.1A Active CN117786373B (en) | 2024-02-28 | 2024-02-28 | Equipment operation diagnosis system based on big data corrugated paper processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117786373B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118094107B (en) * | 2024-04-08 | 2024-06-21 | 北京华力兴科技发展有限责任公司 | Abnormal data detection method, abnormal data diagnostic device, and radiation thickness gauge |
| CN118037280B (en) * | 2024-04-12 | 2024-06-11 | 山东鑫林纸制品有限公司 | Corrugated paper production line maintenance and fault diagnosis system |
| CN118394046A (en) * | 2024-05-08 | 2024-07-26 | 北京东方昊为工业装备有限公司 | Intelligent overhead operation monitoring system and method |
| CN118154174B (en) * | 2024-05-11 | 2024-08-20 | 青岛研博数据信息技术有限公司 | Intelligent operation and maintenance cloud platform for industrial equipment |
| CN118194150B (en) * | 2024-05-13 | 2024-08-02 | 无锡冠亚恒温制冷技术有限公司 | Remote monitoring and fault prediction system for water chiller |
| CN118209830B (en) * | 2024-05-14 | 2024-07-23 | 山东博通节能科技有限公司 | Intelligent monitoring method and system for cable partial discharge abnormality |
| CN118439292B (en) * | 2024-05-28 | 2024-10-01 | 浙江维安支护科技股份有限公司 | Control method and system of automatic material hoister for disc buckle |
| CN118312911B (en) * | 2024-06-07 | 2024-08-16 | 深能智慧能源科技有限公司 | Diagnostic fault early warning system and method based on data mining |
| CN118437494B (en) * | 2024-07-09 | 2024-10-11 | 杭州百瑞特饲料科技有限公司 | Optimized control method and system for feed crushing equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111025149A (en) * | 2019-12-13 | 2020-04-17 | 株洲三新包装技术有限公司 | System and method for automatic fault diagnosis of corrugated paper printing machine |
| CN111201494A (en) * | 2017-09-07 | 2020-05-26 | 弗斯伯股份公司 | Predictive diagnostic method for corrugated board production plant |
| CN114880575A (en) * | 2022-05-27 | 2022-08-09 | 清华大学深圳国际研究生院 | User behavior prediction method, device, equipment, medium and product |
| CN116717437A (en) * | 2023-07-19 | 2023-09-08 | 西安热工研究院有限公司 | Wind turbine generator system fault monitoring method and system |
| CN116861343A (en) * | 2023-07-10 | 2023-10-10 | 广东德尔智慧科技股份有限公司 | Bearing fault diagnosis method |
| CN117032165A (en) * | 2023-08-08 | 2023-11-10 | 贵州理工学院 | Industrial equipment fault diagnosis method |
| KR102618023B1 (en) * | 2022-12-23 | 2023-12-27 | 앤츠이엔씨 주식회사 | Failure prediction diagnosis system and method through pattern analysis according to failure type |
| CN117422935A (en) * | 2023-12-13 | 2024-01-19 | 深圳市鑫思诚科技有限公司 | Motorcycle fault non-contact diagnosis method and system |
| CN117436004A (en) * | 2023-12-20 | 2024-01-23 | 深圳市鸿明机电有限公司 | Motor performance real-time monitoring system and method |
| CN117457192A (en) * | 2023-12-22 | 2024-01-26 | 深圳市健怡康医疗器械科技有限公司 | Intelligent remote diagnosis method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230214693A1 (en) * | 2021-12-31 | 2023-07-06 | Bmc Software, Inc. | Sequence prediction explanation using causal chain extraction based on neural network attributions |
-
2024
- 2024-02-28 CN CN202410218965.1A patent/CN117786373B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111201494A (en) * | 2017-09-07 | 2020-05-26 | 弗斯伯股份公司 | Predictive diagnostic method for corrugated board production plant |
| CN111025149A (en) * | 2019-12-13 | 2020-04-17 | 株洲三新包装技术有限公司 | System and method for automatic fault diagnosis of corrugated paper printing machine |
| CN114880575A (en) * | 2022-05-27 | 2022-08-09 | 清华大学深圳国际研究生院 | User behavior prediction method, device, equipment, medium and product |
| KR102618023B1 (en) * | 2022-12-23 | 2023-12-27 | 앤츠이엔씨 주식회사 | Failure prediction diagnosis system and method through pattern analysis according to failure type |
| CN116861343A (en) * | 2023-07-10 | 2023-10-10 | 广东德尔智慧科技股份有限公司 | Bearing fault diagnosis method |
| CN116717437A (en) * | 2023-07-19 | 2023-09-08 | 西安热工研究院有限公司 | Wind turbine generator system fault monitoring method and system |
| CN117032165A (en) * | 2023-08-08 | 2023-11-10 | 贵州理工学院 | Industrial equipment fault diagnosis method |
| CN117422935A (en) * | 2023-12-13 | 2024-01-19 | 深圳市鑫思诚科技有限公司 | Motorcycle fault non-contact diagnosis method and system |
| CN117436004A (en) * | 2023-12-20 | 2024-01-23 | 深圳市鸿明机电有限公司 | Motor performance real-time monitoring system and method |
| CN117457192A (en) * | 2023-12-22 | 2024-01-26 | 深圳市健怡康医疗器械科技有限公司 | Intelligent remote diagnosis method and system |
Non-Patent Citations (7)
| Title |
|---|
| Multi-Variate, Recurrent Neural Network in a Short-Term Time-Series Substation Demand Forecasting;Ariel B. Suan等;2022 Saudi Arabia Smart Grid (SASG);20230810;全文 * |
| 从信息学的角度分析复杂网络链路预测;王慧;乐孜纯;龚轩;武玉坤;左浩;;小型微型计算机系统;20200215(02);全文 * |
| 基于RNN-LSTM的磨矿系统故障诊断技术;曲星宇;曾鹏;李俊鹏;;信息与控制;20190415(02);全文 * |
| 基于局部密度的加权一类支持向量机算法及其在涡轴发动机故障检测中的应用;黄功;赵永平;谢云龙;;计算机应用;20200331(03);全文 * |
| 基于注意力机制BiLSTM的设备智能故障诊断方法;王太勇;王廷虎;王鹏;乔卉卉;徐明达;;天津大学学报(自然科学与工程技术版);20200427(06);全文 * |
| 基于知识的单面瓦楞纸板机生产故障诊断系统的研究;陈兴国;包装工程;19950815(04);全文 * |
| 数据融合算法在畜禽舍环境监测系统中的应用;邵林;刘淑霞;霍晓静;赵晓顺;王辉;;农机化研究;20130801(08);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117786373A (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117786373B (en) | Equipment operation diagnosis system based on big data corrugated paper processing | |
| Kang et al. | Machine learning applications in production lines: A systematic literature review | |
| Arena et al. | A novel decision support system for managing predictive maintenance strategies based on machine learning approaches | |
| KR102618023B1 (en) | Failure prediction diagnosis system and method through pattern analysis according to failure type | |
| Ouadah et al. | Selecting an appropriate supervised machine learning algorithm for predictive maintenance | |
| Kundu et al. | Multiple failure behaviors identification and remaining useful life prediction of ball bearings | |
| US8732100B2 (en) | Method and apparatus for event detection permitting per event adjustment of false alarm rate | |
| CN118154174B (en) | Intelligent operation and maintenance cloud platform for industrial equipment | |
| WO2002031613A2 (en) | System and method for monitoring process quality control | |
| KR20190062739A (en) | Method, algorithm and device for Data analytics for predictive maintenance using multiple sensors | |
| EP3296822B1 (en) | Model-plant mismatch detection using model parameter data clustering for paper machines or other systems | |
| US12007745B2 (en) | Apparatus for predicting equipment damage | |
| EP4033321B1 (en) | System and method for performance and health monitoring to optimize operation of a pulverizer mill | |
| Riazi et al. | Detecting the onset of machine failure using anomaly detection methods | |
| da Silva Arantes et al. | A novel unsupervised method for anomaly detection in time series based on statistical features for industrial predictive maintenance | |
| CN116705272A (en) | Comprehensive evaluation method for equipment health state based on multidimensional diagnosis | |
| Becherer et al. | Intelligent choice of machine learning methods for predictive maintenance of intelligent machines | |
| US20240126244A1 (en) | Information Processing Method, Information Processing Apparatus, Molding Machine System and Non-Transitory Computer Readable Recording Medium | |
| Lee et al. | Intelligent factory agents with predictive analytics for asset management | |
| Efeoğlu et al. | Machine learning for predictive maintenance: support vector machines and different kernel functions | |
| KR102334590B1 (en) | Fault detection prediction system for manufacturing machines using artificial neural networks | |
| Husain et al. | Predictive maintenance of single phase ac motor using iot sensor data and machine learning (simple linear regression and multiple linear regression algorithms) | |
| Vicêncio et al. | An intelligent predictive maintenance approach based on end-of-line test logfiles in the automotive industry | |
| KR102684101B1 (en) | System for providing facility predictive maintenance and tool breakage predict service using intelligent robotic process automation | |
| Giannoulidis et al. | Engineering and evaluating an unsupervised predictive maintenance solution: a cold-forming press case-study |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |