CN112199898A - Instrument and equipment fault prediction and health management algorithm based on big data - Google Patents
Instrument and equipment fault prediction and health management algorithm based on big data Download PDFInfo
- Publication number
- CN112199898A CN112199898A CN202011254305.7A CN202011254305A CN112199898A CN 112199898 A CN112199898 A CN 112199898A CN 202011254305 A CN202011254305 A CN 202011254305A CN 112199898 A CN112199898 A CN 112199898A
- Authority
- CN
- China
- Prior art keywords
- data
- simulation
- equipment
- storage
- database
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24552—Database cache management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides an instrument and equipment fault prediction and health management algorithm based on big data, which comprises the following steps: acquiring signal data of instrument equipment through a stress sensor, an acoustic sensor and a vibration sensor, transmitting the signal data to an equipment management center, and transmitting real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface; preprocessing the signal data and the operation parameters sent to the equipment management center to obtain preprocessed data; carrying out simulation analysis on the digital twin equipment by using a simulation tool according to the processed data to obtain simulation data; extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics; and predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feed-forward neural network algorithm according to the data characteristics to obtain a prediction result.
Description
Technical Field
The invention provides an instrument and equipment fault prediction and health management algorithm based on big data, and belongs to the technical field of equipment management.
Background
In the instrument and equipment management process, enterprises mostly maintain major repair, minor repair, spot inspection and the like of equipment according to experience, and some enterprises predict equipment faults by using a major data technology, firstly collect static data and dynamic data of the equipment, then establish an association relation between the equipment state and various data, and finally predict the faults through the obtained real-time data, wherein the methods are based on historical data and experience. The existing equipment management method still has the following problems:
1. the difference between the prediction result of the fault prediction method based on historical data and experience and the actual fault occurrence situation is large, so that the spare part stock is excessive or insufficient;
2. after the instrument equipment breaks down, the achievement rate of the production task is influenced, and the comprehensive utilization rate of the equipment is still low.
Disclosure of Invention
The invention provides an instrument and equipment fault prediction and health management algorithm based on big data, which is used for solving the problems that the difference between a fault prediction result and the actual fault occurrence situation is larger and the comprehensive utilization rate of equipment is lower, and adopts the following technical scheme:
a big-data based instrument failure prediction and health management algorithm, the method comprising:
acquiring signal data of instrument equipment through a stress sensor, an acoustic sensor and a vibration sensor, transmitting the signal data to an equipment management center, and transmitting real-time operation parameters of the instrument equipment to the equipment management center through an interface;
preprocessing the signal data and the operation parameters sent to the equipment management center to obtain preprocessed data;
carrying out simulation analysis on the digital twin equipment by using a simulation tool according to the processed data to obtain simulation data;
extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics;
and predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feed-forward neural network algorithm according to the data characteristics to obtain a prediction result.
Further, the device management center stores the signal data and the operation parameters through a Kafka tool, and the method comprises the following steps:
receiving data through a Kafka tool, and sending the data to a database cache region through a Kafka message bus;
after receiving the data, the database cache region classifies the signal data and the operation parameters based on data types, and sorts the signal data and the operation parameters in respective corresponding data types according to the time sequence of data generation to obtain a data set;
the database cache region extracts the storage space value occupied by the data of the data set, and sends a storage request and the storage space value required by each data set to each database according to the request sending time interval;
after receiving a storage request and a storage space value required by each data set, each database detects whether the database completes the last data storage action, and if the database completes the data storage action, the time and the storage space residual quantity value used by the database for last data storage are sent to the database cache region; if the data storage action is not finished, sending a storage suspension request to the data cache region, and after the data storage action is finished, storing the data for a long time;
the database cache region takes the database corresponding to the returned storage space surplus as a candidate storage database, extracts the storage space surplus returned last time by the candidate storage database, and compares the storage space surplus currently returned by the candidate storage database with the storage space surplus returned last time to obtain a surplus difference value;
canceling the candidate qualification of the candidate storage database when the residual difference exceeds a first preset difference threshold; when the residual difference value exceeds a second preset difference threshold value and does not exceed a first preset difference threshold value, or the residual difference value does not exceed a second preset difference threshold value, reserving the candidate qualification of the candidate storage database;
the data buffer area compares the number of the data sets with the number of the candidate storage databases, and stores data according to the comparison result and the storage rule;
after receiving the time used for storing the data of the same batch fed back by all the databases, the data buffer adjusts the request sending time interval according to the following formula:
wherein T represents a transmission request time interval; m represents the number of the databases; t is0Indicating an initial default value for the transmission request interval; t isiThe data storage time for storing the data in the same batch fed back by the ith database is represented; t ismaxThe maximum time value used by the database for storing data in the same batch of storage processes is represented; t isminAnd the minimum time value used by the database for data storage in the same batch of storage processes is represented.
Further, the storage rule is:
when the number of the data sets is smaller than that of the candidate database, screening out the candidate databases with the number corresponding to the number of the data sets according to the sequence from small to large of the residual difference, and sequentially storing each data set into the candidate databases arranged according to the sequence from small to large of the residual difference according to the sequence from large to small of the storage space occupied by the data;
when the number of the data sets is larger than that of the candidate storage databases, sequentially arranging the data sets according to the sequence of the storage space occupied by the data from large to small to obtain a data set sequence; storing the first A data sets in the data set sequence into a candidate database of which the residual difference value does not exceed a second preset difference threshold value, storing the B data sets in sequence into a candidate database of which the residual difference value exceeds the second preset difference threshold value and does not exceed a first preset difference threshold value, and storing the C data sets into a candidate database of which the residual difference value is minimum after finishing storing the A data sets and the B data sets;
wherein, A corresponds to the number of the candidate databases of which the residual difference value does not exceed a second preset difference threshold, B corresponds to the number of the candidate databases of which the residual difference value exceeds the second preset difference threshold and does not exceed a first preset difference threshold, and C corresponds to the difference value between the number of the data sets and the number of the candidate storage databases.
Further, the signal data and the operation parameters sent to the equipment management center are preprocessed to obtain preprocessed data, and the preprocessing comprises the following steps:
checking the data type of the signal data and the data corresponding to the operating parameters;
carrying out data type verification on the signal data and the data corresponding to the operating parameters;
performing data cleaning on the signal data and the operation parameters after the data type checking and the data type checking are completed to obtain cleaned data;
and carrying out data integration, data transformation and data specification processing on the cleaned data to obtain preprocessed data.
Further, performing simulation analysis on the digital twin device by using a simulation tool according to the processed data to obtain simulation data, wherein the simulation data comprises:
carrying out simulation analysis on the digital twin equipment by using an APDLANSYS virtual simulation method aiming at the processed data, and simulating the running condition of the instrument equipment in real time to obtain simulated data;
and performing consistency comparison on the simulation data and the operation data of the instrument equipment in real time, and keeping consistency between the simulation data and the operation data of the instrument equipment.
Further, extracting data features according to the simulation data and the actual data of the instrument and equipment to obtain the data features, including:
performing data feature extraction on the simulation data and actual data of the instrument equipment by using a Spark tool;
and inputting the data extracted by the Spark tool into the self-encoder, and processing the input data by using the self-encoder to acquire data characteristics.
Further, the device management center includes:
the data acquisition and transmission module is used for acquiring signal data of instrument equipment through the stress sensor, the acoustic sensor and the vibration sensor, transmitting the signal data to an equipment management center, and simultaneously transmitting real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface;
the device comprises a preprocessing module, a data processing module and a data processing module, wherein the preprocessing module is used for preprocessing signal data and operation parameters sent to a device management center to obtain preprocessed data;
the simulation module is used for carrying out simulation analysis on the digital twin equipment by utilizing a simulation tool aiming at the processed data to obtain simulation data;
the characteristic extraction module is used for extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics;
and the prediction module is used for predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feedforward neural network algorithm according to the data characteristics to obtain a prediction result.
Further, the preprocessing module comprises:
the data type checking module is used for checking the data types of the data corresponding to the signal data and the operation parameters;
the data type checking module is used for carrying out data type checking on the signal data and the data corresponding to the operating parameters;
the data cleaning module is used for cleaning the signal data and the operation parameters which are used for finishing the data type checking and the data type checking to obtain the cleaned data;
and the data processing module is used for carrying out data integration, data transformation and data protocol processing on the cleaned data to obtain the preprocessed data.
Further, the simulation module includes:
the simulation analysis module is used for performing simulation analysis on the digital twin equipment according to the processed data by utilizing an APDLANSYS virtual simulation method, simulating the running condition of the instrument equipment in real time and obtaining simulated data;
and the consistency comparison module is used for carrying out consistency comparison on the simulation data and the operation data of the instrument equipment in real time and keeping consistency between the simulation data and the operation data of the instrument equipment.
Further, the feature extraction module comprises:
the data feature acquisition module is used for extracting data features of the simulation data and actual data of the instrument equipment by utilizing a Spark tool;
the sending module is used for inputting the data extracted by the Spark tool into the self-encoder;
and the self-encoder is used for processing the input data and acquiring data characteristics.
The invention has the beneficial effects that:
compared with the traditional instrument and equipment health management and fault diagnosis, the instrument and equipment fault prediction and health management algorithm based on the big data can realize the real-time interaction and omnibearing state comparison of physical and virtual equipment, simulate the physical state of the equipment in real time, capture more comprehensive equipment operation characteristics, diagnose and predict faults more accurately and verify maintenance strategies more accurately. Therefore, the demand of spare parts can be predicted more accurately, and shortage or excess can be avoided. The more targeted maintenance strategy can reduce the fault downtime of the equipment and provide the comprehensive utilization rate of the equipment.
Drawings
FIG. 1 is a flow chart of an algorithm according to the present invention;
FIG. 2 is a schematic diagram of the algorithm of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The embodiment of the invention provides an instrument and equipment fault prediction and health management algorithm based on big data, and as shown in fig. 1 and fig. 2, the method comprises the following steps:
s1, collecting signal data of instrument equipment through a stress sensor, an acoustic sensor and a vibration sensor, sending the signal data to an equipment management center, and sending real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface;
s2, preprocessing the signal data and the operation parameters sent to the equipment management center to obtain preprocessed data;
s3, carrying out simulation analysis on the digital twin equipment by using a simulation tool according to the processed data to obtain simulation data;
s4, extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics;
and S5, predicting the loss and the fault of the instrument and/or the key components of the instrument by utilizing a feed-forward neural network algorithm according to the data characteristics to obtain a prediction result.
The working principle of the technical scheme is as follows: firstly, data acquisition is carried out, various sensors such as a stress sensor, an acoustic sensor, a vibration sensor and the like are additionally arranged on instrument equipment, and signal data collected by the sensors are transmitted to an equipment management system in real time; the operation data of the equipment, including power, current, feed amount, work task, work duration and the like, is acquired through an interface of the equipment, and the acquired information is transmitted to an equipment management system in real time. Then, data processing is performed on the acquired data, namely, the acquired data is preprocessed, wherein the preprocessing comprises data type checking, data cleaning, data transformation and the like, and the preprocessing process is realized based on spark streaming. The data after preprocessing can be used for simulation analysis. Finally, carrying out simulation analysis according to the preprocessed data, which specifically comprises the following steps: and performing simulation analysis on the digital twin equipment by using ANSYS, wherein the simulation analysis data is derived from real-time equipment static and dynamic data, the running condition of the physical equipment can be simulated in real time by using a simulation tool, and the running condition is compared with the running data consistency of the physical equipment in real time, so that the reliability of a simulation result is ensured. Then, data feature extraction is carried out according to the simulation data and the actual data of the physical equipment, and the feature extraction can be completed by applying methods such as continuous wavelet transformation, maximum stress simulation and the like during extraction. Meanwhile, the extraction process of the data features is divided into a plurality of stages, the first stage is to use a Spark tool to extract the features, and the second stage is to extract the features based on a self-encoder, so that the dimension reduction of the data, the redundancy reduction and the deeper understanding of the data are realized. And finally, predicting the fault and the service life according to the simulation data and the data characteristics extracted from the actual data of the physical equipment.
The effect of the above technical scheme is as follows: compared with the traditional instrument and equipment health management and fault diagnosis, the instrument and equipment fault prediction and health management algorithm based on the big data can realize the real-time interaction and omnibearing state comparison of physical and virtual equipment, simulate the physical state of the equipment in real time, capture more comprehensive equipment operation characteristics, diagnose and predict faults more accurately and verify maintenance strategies more accurately. Therefore, the demand of spare parts can be predicted more accurately, and shortage or excess can be avoided. The more targeted maintenance strategy can reduce the fault downtime of the equipment and provide the comprehensive utilization rate of the equipment.
In an embodiment of the present invention, the device management center stores the signal data and the operation parameters by using a Kafka tool, including:
step 1, receiving data through a Kafka tool, and sending the data to a database cache region through a Kafka message bus;
step 2, after receiving the data, the database cache region classifies the signal data and the operation parameters based on data types, and sorts the signal data and the operation parameters in respective corresponding data types according to the time sequence of data generation to obtain a data set;
step 3, the database cache region extracts the storage space value occupied by the data of the data set, and sends a storage request and the storage space value required by each data set to each database according to the request sending time interval;
step 4, after receiving the storage request and the storage space value required by each data set, each database detects whether the database completes the last data storage action, and if the data storage action is completed, the time and the storage space residual quantity value used by the database for the last data storage are sent to the database cache region; if the data storage action is not finished, sending a storage suspension request to the data cache region, and after the data storage action is finished, storing the data for a long time;
step 5, the database cache region takes the database corresponding to the returned storage space surplus as a candidate storage database, extracts the storage space surplus returned last time by the candidate storage database, and compares the storage space surplus currently returned by the candidate storage database with the storage space surplus returned last time to obtain a surplus difference value;
step 6, when the residual difference value exceeds a first preset difference value threshold value, canceling the candidate qualification of the candidate storage database; when the residual difference value exceeds a second preset difference threshold value and does not exceed a first preset difference threshold value, or the residual difference value does not exceed a second preset difference threshold value, reserving the candidate qualification of the candidate storage database;
step 7, the data buffer area compares the number of the data sets with the number of the candidate storage databases, and data storage is carried out according to the comparison result and the storage rule;
step 8, after receiving the time used for storing the data of the same batch fed back by all the databases, the data buffer adjusts the request sending time interval according to the following formula:
wherein T represents a transmission request time interval; m represents the number of the databases; t is0Indicating an initial default value for the transmission request interval; t isiThe data storage time for storing the data in the same batch fed back by the ith database is represented; t ismaxThe maximum time value used by the database for storing data in the same batch of storage processes is represented; t isminAnd the minimum time value used by the database for data storage in the same batch of storage processes is represented.
Wherein the storage rule is:
when the number of the data sets is smaller than that of the candidate database, screening out the candidate databases with the number corresponding to the number of the data sets according to the sequence from small to large of the residual difference, and sequentially storing each data set into the candidate databases arranged according to the sequence from small to large of the residual difference according to the sequence from large to small of the storage space occupied by the data;
when the number of the data sets is larger than that of the candidate storage databases, sequentially arranging the data sets according to the sequence of the storage space occupied by the data from large to small to obtain a data set sequence; storing the first A data sets in the data set sequence into a candidate database of which the residual difference value does not exceed a second preset difference threshold value, storing the B data sets in sequence into a candidate database of which the residual difference value exceeds the second preset difference threshold value and does not exceed a first preset difference threshold value, and storing the C data sets into a candidate database of which the residual difference value is minimum after finishing storing the A data sets and the B data sets;
wherein, A corresponds to the number of the candidate databases of which the residual difference value does not exceed a second preset difference threshold, B corresponds to the number of the candidate databases of which the residual difference value exceeds the second preset difference threshold and does not exceed a first preset difference threshold, and C corresponds to the difference value between the number of the data sets and the number of the candidate storage databases.
The effect of the above technical scheme is as follows: by the method and the storage rule for data storage, the saturation rate of the storage space can be effectively reduced, meanwhile, the balance of the storage space surplus among the databases can be improved, and the problem that the data is unreasonably used due to the fact that the storage occupation difference among the data storage spaces is too large is avoided. Meanwhile, the data storage efficiency can be effectively improved through the arrangement of the database cache region and the storage steps, the timeliness and the speed of data storage are improved in the dynamic data storage process of big data, and the problems of reduction of the storage efficiency caused by overhigh data generation frequency and overhigh storage requirement frequency and data overstock caused by overhigh data waiting amount are effectively avoided. Meanwhile, through the self-adaptive adjustment of the sending time interval and the setting of the adjustment formula, the data backlog in the buffer area of the database can be effectively reduced, the data entry amount and the data output amount (namely the data amount sent to the database) in the buffer area of the database are kept highly balanced and reasonable, the problem of data backlog in the buffer area of the database is effectively reduced, and the data storage efficiency is improved. Meanwhile, the sending time interval of the data storage is adjusted according to the data storage time of the database, so that the balance between the data storage and the data storage sending can be effectively improved according to the actual data storage situation.
In an embodiment of the present invention, preprocessing signal data and operation parameters sent to an equipment management center to obtain preprocessed data includes:
s201, checking the data type of the signal data and the data corresponding to the operation parameters;
s202, carrying out data type verification on the signal data and the data corresponding to the operating parameters;
s203, performing data cleaning on the signal data and the operation parameters after the data type checking and the data type checking are completed, and obtaining cleaned data;
and S204, performing data integration, data transformation and data specification processing on the cleaned data to obtain preprocessed data.
The working principle of the technical scheme is as follows: firstly, checking the data type of the signal data and the data corresponding to the operation parameters; then, carrying out data type verification on the signal data and the data corresponding to the operating parameters; then, carrying out data cleaning on the signal data and the operation parameters after the data type checking and the data type checking are finished, and obtaining cleaned data; and finally, performing data integration, data transformation and data specification processing on the cleaned data to obtain preprocessed data.
The effect of the above technical scheme is as follows: the method effectively reduces noise signals in the signal data and the operation parameters, improves the accuracy of subsequent processing of the signal data and the operation parameters, further improves the accuracy of equipment fault prediction, and reduces the difference between a fault prediction result and the actual fault occurrence condition.
In an embodiment of the present invention, performing simulation analysis on the digital twin device by using a simulation tool with respect to the processed data to obtain simulation data, includes:
s301, carrying out simulation analysis on the digital twin equipment according to the processed data by utilizing an APDLANSYS virtual simulation method, and simulating the running condition of the instrument equipment in real time to obtain simulated data;
s302, performing consistency comparison on the simulation data and the operation data of the instrument in real time, and keeping consistency between the simulation data and the operation data of the instrument.
The working principle of the technical scheme is as follows: firstly, carrying out simulation analysis on the digital twin equipment by using an APDLANSYS virtual simulation method aiming at the processed data, and simulating the running condition of the instrument equipment in real time to obtain simulated data; and then, carrying out consistency comparison on the simulation data and the operation data of the instrument equipment in real time, and keeping consistency between the simulation data and the operation data of the instrument equipment.
The effect of the above technical scheme is as follows: the reliability of the simulation result is effectively improved, the accuracy of equipment fault prediction is further improved, and the difference between the fault prediction result and the actual fault occurrence condition is reduced.
According to an embodiment of the present invention, extracting data features according to the simulation data and the actual data of the instrument and device to obtain data features includes:
s401, data feature extraction is carried out on the simulation data and actual data of the instrument equipment by using a Spark tool;
s402, inputting the data extracted by the Spark tool into a self-encoder, and processing the input data by the self-encoder to acquire data characteristics.
The working principle of the technical scheme is as follows: firstly, performing data feature extraction on the simulation data and actual data of the instrument equipment by using a Spark tool; then, the data extracted by the Spark tool is input into the self-encoder, and the self-encoder is used for processing the input data to acquire data characteristics. The Spark tool performs targeted extraction on different characteristics of data in modes of continuous wavelet transformation, kurtosis index acquisition, effective value acquisition, maximum stress simulation, energy simulation, stress cycle times and the like.
The effect of the above technical scheme is as follows: by means of the data feature extraction, accuracy and feature comprehensiveness of data feature extraction can be effectively improved, accuracy of equipment fault prediction is further improved, and difference between a fault prediction result and an actual fault occurrence condition is reduced. Meanwhile, through the feature extraction processing of the two stages, the dimension reduction of the data can be effectively realized, the redundancy is reduced, the equipment management center can deeply know the data in the data processing process, and the prediction accuracy of the equipment fault can be effectively improved when the fault prediction is carried out by utilizing a feedforward neural network algorithm in the follow-up process.
In an embodiment of the present invention, the device management center includes:
the data acquisition and transmission module is used for acquiring signal data of instrument equipment through the stress sensor, the acoustic sensor and the vibration sensor, transmitting the signal data to an equipment management center, and simultaneously transmitting real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface;
the device comprises a preprocessing module, a data processing module and a data processing module, wherein the preprocessing module is used for preprocessing signal data and operation parameters sent to a device management center to obtain preprocessed data;
the simulation module is used for carrying out simulation analysis on the digital twin equipment by utilizing a simulation tool aiming at the processed data to obtain simulation data;
the characteristic extraction module is used for extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics;
and the prediction module is used for predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feedforward neural network algorithm according to the data characteristics to obtain a prediction result.
The working principle of the technical scheme is as follows: firstly, a data acquisition and transmission module is used for acquiring signal data of instrument equipment through a stress sensor, an acoustic sensor and a vibration sensor, transmitting the signal data to an equipment management center, and simultaneously transmitting real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface; then, preprocessing the signal data and the operation parameters sent to the equipment management center through a preprocessing module to obtain preprocessed data; then, a simulation module is adopted to carry out simulation analysis on the digital twin equipment aiming at the processed data by utilizing a simulation tool to obtain simulation data; then, a feature extraction module is adopted to extract data features according to the simulation data and actual data of the instrument equipment to obtain data features; and finally, predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feed-forward neural network algorithm through a prediction module according to the data characteristics to obtain a prediction result.
The effect of the above technical scheme is as follows: compared with the traditional instrument and equipment health management and fault diagnosis, the instrument and equipment fault prediction and health management algorithm based on the big data can realize the real-time interaction and omnibearing state comparison of physical and virtual equipment, simulate the physical state of the equipment in real time, capture more comprehensive equipment operation characteristics, diagnose and predict faults more accurately and verify maintenance strategies more accurately. Therefore, the demand of spare parts can be predicted more accurately, and shortage or excess can be avoided. The more targeted maintenance strategy can reduce the fault downtime of the equipment and provide the comprehensive utilization rate of the equipment.
In one embodiment of the present invention, the preprocessing module comprises:
the data type checking module is used for checking the data types of the data corresponding to the signal data and the operation parameters;
the data type checking module is used for carrying out data type checking on the signal data and the data corresponding to the operating parameters;
the data cleaning module is used for cleaning the signal data and the operation parameters which are used for finishing the data type checking and the data type checking to obtain the cleaned data;
and the data processing module is used for carrying out data integration, data transformation and data protocol processing on the cleaned data to obtain the preprocessed data.
The working principle of the technical scheme is as follows: firstly, checking the data type of the signal data and the data corresponding to the operating parameters through a data type checking module; then, carrying out data type verification on the signal data and the data corresponding to the operating parameters by using a data type verification module; then, data cleaning is carried out on the signal data and the operation parameters which are subjected to data type checking and data type checking through a data cleaning module, and cleaned data are obtained; and finally, performing data integration, data transformation and data specification processing on the cleaned data by adopting a data processing module to obtain the preprocessed data.
The effect of the above technical scheme is as follows: the method effectively reduces noise signals in the signal data and the operation parameters, improves the accuracy of subsequent processing of the signal data and the operation parameters, further improves the accuracy of equipment fault prediction, and reduces the difference between a fault prediction result and the actual fault occurrence condition.
In one embodiment of the present invention, the simulation module includes:
the simulation analysis module is used for performing simulation analysis on the digital twin equipment according to the processed data by utilizing an APDLANSYS virtual simulation method, simulating the running condition of the instrument equipment in real time and obtaining simulated data;
and the consistency comparison module is used for carrying out consistency comparison on the simulation data and the operation data of the instrument equipment in real time and keeping consistency between the simulation data and the operation data of the instrument equipment.
The working principle of the technical scheme is as follows: firstly, a simulation analysis module carries out simulation analysis of digital twin equipment on the processed data by utilizing an APDL ANSYS virtual simulation method, and simulates the running condition of the instrument equipment in real time to obtain simulated data; and then, carrying out consistency comparison on the simulation data and the operation data of the instrument equipment in real time by using a consistency comparison module, and keeping consistency between the simulation data and the operation data of the instrument equipment.
The effect of the above technical scheme is as follows: the reliability of the simulation result is effectively improved, the accuracy of equipment fault prediction is further improved, and the difference between the fault prediction result and the actual fault occurrence condition is reduced.
In one embodiment of the present invention, the feature extraction module includes:
the data feature acquisition module is used for extracting data features of the simulation data and actual data of the instrument equipment by utilizing a Spark tool;
the sending module is used for inputting the data extracted by the Spark tool into the self-encoder;
and the self-encoder is used for processing the input data and acquiring data characteristics.
The working principle of the technical scheme is as follows: firstly, data feature extraction is carried out on the simulation data and actual data of the instrument equipment by a data feature acquisition module through a Spark tool; then, inputting the data extracted by the Spark tool into a self-encoder by adopting a sending module; and finally, processing the input data through an autoencoder to obtain data characteristics.
The effect of the above technical scheme is as follows: by means of the data feature extraction, accuracy and feature comprehensiveness of data feature extraction can be effectively improved, accuracy of equipment fault prediction is further improved, and difference between a fault prediction result and an actual fault occurrence condition is reduced. Meanwhile, through the feature extraction processing of the two stages, the dimension reduction of the data can be effectively realized, the redundancy is reduced, the equipment management center can deeply know the data in the data processing process, and the prediction accuracy of the equipment fault can be effectively improved when the fault prediction is carried out by utilizing a feedforward neural network algorithm in the follow-up process.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. An instrument equipment fault prediction and health management algorithm based on big data, characterized in that the method comprises:
acquiring signal data of instrument equipment through a stress sensor, an acoustic sensor and a vibration sensor, transmitting the signal data to an equipment management center, and transmitting real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface;
preprocessing the signal data and the operation parameters sent to the equipment management center to obtain preprocessed data;
carrying out simulation analysis on the digital twin equipment by using a simulation tool according to the processed data to obtain simulation data;
extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics;
and predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feed-forward neural network algorithm according to the data characteristics to obtain a prediction result.
2. The algorithm of claim 1, wherein the device management center stores the signal data and the operating parameters via a Kafka tool, comprising:
receiving data through a Kafka tool, and sending the data to a database cache region through a Kafka message bus;
after receiving the data, the database cache region classifies the signal data and the operation parameters based on data types, and sorts the signal data and the operation parameters in respective corresponding data types according to the time sequence of data generation to obtain a data set;
the database cache region extracts the storage space value occupied by the data of the data set, and sends a storage request and the storage space value required by each data set to each database according to the request sending time interval;
after receiving a storage request and a storage space value required by each data set, each database detects whether the database completes the last data storage action, and if the database completes the data storage action, the time and the storage space residual quantity value used by the database for last data storage are sent to the database cache region; if the data storage action is not finished, sending a storage suspension request to the data cache region, and after the data storage action is finished, storing the data for a long time;
the database cache region takes the database corresponding to the returned storage space surplus as a candidate storage database, extracts the storage space surplus returned last time by the candidate storage database, and compares the storage space surplus currently returned by the candidate storage database with the storage space surplus returned last time to obtain a surplus difference value;
canceling the candidate qualification of the candidate storage database when the residual difference exceeds a first preset difference threshold; when the residual difference value exceeds a second preset difference threshold value and does not exceed a first preset difference threshold value, or the residual difference value does not exceed a second preset difference threshold value, reserving the candidate qualification of the candidate storage database;
the data buffer area compares the number of the data sets with the number of the candidate storage databases, and stores data according to the comparison result and the storage rule;
after receiving the time used for storing the data of the same batch fed back by all the databases, the data buffer adjusts the request sending time interval according to the following formula:
wherein T represents a transmission request time interval; m represents the number of the databases; t is0Indicating an initial default value for the transmission request interval; t isiThe data storage time for storing the data in the same batch fed back by the ith database is represented; t ismaxRepresenting the same batch of storage processes, the database is enteredA maximum time value for line data storage; t isminAnd the minimum time value used by the database for data storage in the same batch of storage processes is represented.
3. The algorithm of claim 2, wherein the storage rule is:
when the number of the data sets is smaller than that of the candidate database, screening out the candidate databases with the number corresponding to the number of the data sets according to the sequence from small to large of the residual difference, and sequentially storing each data set into the candidate databases arranged according to the sequence from small to large of the residual difference according to the sequence from large to small of the storage space occupied by the data;
when the number of the data sets is larger than that of the candidate storage databases, sequentially arranging the data sets according to the sequence of the storage space occupied by the data from large to small to obtain a data set sequence; storing the first A data sets in the data set sequence into a candidate database of which the residual difference value does not exceed a second preset difference threshold value, storing the B data sets in sequence into a candidate database of which the residual difference value exceeds the second preset difference threshold value and does not exceed a first preset difference threshold value, and storing the C data sets into a candidate database of which the residual difference value is minimum after finishing storing the A data sets and the B data sets;
wherein, A corresponds to the number of the candidate databases of which the residual difference value does not exceed a second preset difference threshold, B corresponds to the number of the candidate databases of which the residual difference value exceeds the second preset difference threshold and does not exceed a first preset difference threshold, and C corresponds to the difference value between the number of the data sets and the number of the candidate storage databases.
4. The algorithm of claim 1, wherein pre-processing the signal data and the operating parameters sent to the device management center to obtain pre-processed data comprises:
checking the data type of the signal data and the data corresponding to the operating parameters;
carrying out data type verification on the signal data and the data corresponding to the operating parameters;
performing data cleaning on the signal data and the operation parameters after the data type checking and the data type checking are completed to obtain cleaned data;
and carrying out data integration, data transformation and data specification processing on the cleaned data to obtain preprocessed data.
5. The algorithm of claim 1, wherein performing simulation analysis of the digital twin device on the processed data using a simulation tool to obtain simulation data comprises:
carrying out simulation analysis on the digital twin equipment by using an APDL ANSYS virtual simulation method aiming at the processed data, and simulating the running condition of the instrument equipment in real time to obtain simulated data;
and performing consistency comparison on the simulation data and the operation data of the instrument equipment in real time, and keeping consistency between the simulation data and the operation data of the instrument equipment.
6. The algorithm of claim 1, wherein performing data feature extraction according to the simulation data and actual data of the instrument device to obtain data features comprises:
performing data feature extraction on the simulation data and actual data of the instrument equipment by using a Spark tool;
and inputting the data extracted by the Spark tool into the self-encoder, and processing the input data by using the self-encoder to acquire data characteristics.
7. The algorithm of claim 1, wherein the device management center comprises:
the data acquisition and transmission module is used for acquiring signal data of instrument equipment through the stress sensor, the acoustic sensor and the vibration sensor, transmitting the signal data to an equipment management center, and simultaneously transmitting real-time operation parameters of the instrument equipment to the equipment management center through a machine tool interface;
the device comprises a preprocessing module, a data processing module and a data processing module, wherein the preprocessing module is used for preprocessing signal data and operation parameters sent to a device management center to obtain preprocessed data;
the simulation module is used for carrying out simulation analysis on the digital twin equipment by utilizing a simulation tool aiming at the processed data to obtain simulation data;
the characteristic extraction module is used for extracting data characteristics according to the simulation data and actual data of the instrument equipment to obtain data characteristics;
and the prediction module is used for predicting the loss and the fault of the instrument and/or the key component of the instrument by utilizing a feedforward neural network algorithm according to the data characteristics to obtain a prediction result.
8. The algorithm of claim 6, wherein the preprocessing module comprises:
the data type checking module is used for checking the data types of the data corresponding to the signal data and the operation parameters;
the data type checking module is used for carrying out data type checking on the signal data and the data corresponding to the operating parameters;
the data cleaning module is used for cleaning the signal data and the operation parameters which are used for finishing the data type checking and the data type checking to obtain the cleaned data;
and the data processing module is used for carrying out data integration, data transformation and data protocol processing on the cleaned data to obtain the preprocessed data.
9. The algorithm of claim 6, wherein the simulation module comprises:
the simulation analysis module is used for performing simulation analysis on the digital twin equipment according to the processed data by utilizing an APDL ANSYS virtual simulation method, simulating the running condition of the instrument equipment in real time and obtaining simulated data;
and the consistency comparison module is used for carrying out consistency comparison on the simulation data and the operation data of the instrument equipment in real time and keeping consistency between the simulation data and the operation data of the instrument equipment.
10. The algorithm of claim 6, wherein the feature extraction module comprises:
the data feature acquisition module is used for extracting data features of the simulation data and actual data of the instrument equipment by utilizing a Spark tool;
the sending module is used for inputting the data extracted by the Spark tool into the self-encoder;
and the self-encoder is used for processing the input data and acquiring data characteristics.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011254305.7A CN112199898B (en) | 2020-11-11 | 2020-11-11 | Instrument and equipment fault prediction and health management method based on big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011254305.7A CN112199898B (en) | 2020-11-11 | 2020-11-11 | Instrument and equipment fault prediction and health management method based on big data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112199898A true CN112199898A (en) | 2021-01-08 |
CN112199898B CN112199898B (en) | 2021-06-15 |
Family
ID=74033416
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011254305.7A Active CN112199898B (en) | 2020-11-11 | 2020-11-11 | Instrument and equipment fault prediction and health management method based on big data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112199898B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114519437A (en) * | 2022-02-18 | 2022-05-20 | 江苏兴教科技有限公司 | Cloud-based micro-service method and system for fault diagnosis analysis and repair reporting |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103856558A (en) * | 2014-01-22 | 2014-06-11 | 北京京东尚科信息技术有限公司 | Data processing method and device for terminal application |
CN109709389A (en) * | 2018-11-30 | 2019-05-03 | 珠海派诺科技股份有限公司 | For electric instrument distributed mass real time data sampling alarm method and system |
CN109800531A (en) * | 2019-01-31 | 2019-05-24 | 山东大学 | A kind of twin model consistency keeping method of dynamoelectric equipment number |
CN110399642A (en) * | 2019-06-21 | 2019-11-01 | 浙江大学 | It is a kind of for the twin body of number and its construction method of production line and application |
CN110609531A (en) * | 2019-09-27 | 2019-12-24 | 北京航空航天大学 | Workshop scheduling method based on digital twin |
CN110704335A (en) * | 2019-09-03 | 2020-01-17 | 苏州浪潮智能科技有限公司 | Data reading and writing method and device based on asynchronous ring buffer |
CN110900307A (en) * | 2019-11-22 | 2020-03-24 | 北京航空航天大学 | Numerical control machine tool cutter monitoring system driven by digital twin |
CN111008502A (en) * | 2019-11-25 | 2020-04-14 | 北京航空航天大学 | Fault prediction method for complex equipment driven by digital twin |
CN111143414A (en) * | 2019-12-26 | 2020-05-12 | 五八有限公司 | Feedback method and device of cache data, electronic equipment and storage medium |
US20200225655A1 (en) * | 2016-05-09 | 2020-07-16 | Strong Force Iot Portfolio 2016, Llc | Methods, systems, kits and apparatuses for monitoring and managing industrial settings in an industrial internet of things data collection environment |
CN111596604A (en) * | 2020-06-12 | 2020-08-28 | 中国科学院重庆绿色智能技术研究院 | Intelligent fault diagnosis and self-healing control system and method for engineering equipment based on digital twinning |
CN111695734A (en) * | 2020-06-12 | 2020-09-22 | 中国科学院重庆绿色智能技术研究院 | Multi-process planning comprehensive evaluation system and method based on digital twin and deep learning |
-
2020
- 2020-11-11 CN CN202011254305.7A patent/CN112199898B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103856558A (en) * | 2014-01-22 | 2014-06-11 | 北京京东尚科信息技术有限公司 | Data processing method and device for terminal application |
US20200225655A1 (en) * | 2016-05-09 | 2020-07-16 | Strong Force Iot Portfolio 2016, Llc | Methods, systems, kits and apparatuses for monitoring and managing industrial settings in an industrial internet of things data collection environment |
CN109709389A (en) * | 2018-11-30 | 2019-05-03 | 珠海派诺科技股份有限公司 | For electric instrument distributed mass real time data sampling alarm method and system |
CN109800531A (en) * | 2019-01-31 | 2019-05-24 | 山东大学 | A kind of twin model consistency keeping method of dynamoelectric equipment number |
CN110399642A (en) * | 2019-06-21 | 2019-11-01 | 浙江大学 | It is a kind of for the twin body of number and its construction method of production line and application |
CN110704335A (en) * | 2019-09-03 | 2020-01-17 | 苏州浪潮智能科技有限公司 | Data reading and writing method and device based on asynchronous ring buffer |
CN110609531A (en) * | 2019-09-27 | 2019-12-24 | 北京航空航天大学 | Workshop scheduling method based on digital twin |
CN110900307A (en) * | 2019-11-22 | 2020-03-24 | 北京航空航天大学 | Numerical control machine tool cutter monitoring system driven by digital twin |
CN111008502A (en) * | 2019-11-25 | 2020-04-14 | 北京航空航天大学 | Fault prediction method for complex equipment driven by digital twin |
CN111143414A (en) * | 2019-12-26 | 2020-05-12 | 五八有限公司 | Feedback method and device of cache data, electronic equipment and storage medium |
CN111596604A (en) * | 2020-06-12 | 2020-08-28 | 中国科学院重庆绿色智能技术研究院 | Intelligent fault diagnosis and self-healing control system and method for engineering equipment based on digital twinning |
CN111695734A (en) * | 2020-06-12 | 2020-09-22 | 中国科学院重庆绿色智能技术研究院 | Multi-process planning comprehensive evaluation system and method based on digital twin and deep learning |
Non-Patent Citations (4)
Title |
---|
刘丁发等: "《Oracle数据库应用与开发实战》", 31 August 2017 * |
李抒霞 等: "基于SVM的ICMP网络存储隐蔽信道检测", 《信息安全研究》 * |
李福兴 等: "基于数字孪生的船舶预测性维护", 《船舶工程》 * |
程光 等: "《加密流量测量与分析》", 31 December 2018 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114519437A (en) * | 2022-02-18 | 2022-05-20 | 江苏兴教科技有限公司 | Cloud-based micro-service method and system for fault diagnosis analysis and repair reporting |
CN114519437B (en) * | 2022-02-18 | 2022-12-27 | 江苏兴教科技有限公司 | Cloud-based micro-service method and system for fault diagnosis analysis and repair reporting |
Also Published As
Publication number | Publication date |
---|---|
CN112199898B (en) | 2021-06-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107273688B (en) | Method and system for monitoring equipment service life and health degree based on original fault data | |
CN110348615B (en) | Cable line fault probability prediction method based on ant colony optimization support vector machine | |
CN116187593B (en) | Power distribution network fault prediction processing method, device, equipment and storage medium | |
CN112692646B (en) | Intelligent assessment method and device for tool wear state | |
CN102998996A (en) | Airborne real-time fault diagnosis method | |
CN112199898B (en) | Instrument and equipment fault prediction and health management method based on big data | |
CN111930733A (en) | Industrial equipment big data storage system based on monitoring of Internet of things | |
CN110825768A (en) | Remote television exception handling method and system based on cloud analysis | |
CN114330026A (en) | Digital twin system simulation method and device | |
CN111666978B (en) | Intelligent fault early warning system for IT system operation and maintenance big data | |
CN116186936A (en) | Method, system, equipment and medium for determining continuous casting process parameters | |
CN116882790A (en) | Carbon emission equipment management method and system for mine ecological restoration area | |
CN116934043A (en) | Animation design task fine management system | |
CN113030633B (en) | GA-BP neural network-based power distribution network fault big data analysis method and system | |
CN113313365A (en) | Degradation early warning method and device for primary air fan | |
CN108613820A (en) | A kind of online allophone monitoring algorithm for GIS bulk mechanicals defect diagonsis and positioning | |
CN116562599A (en) | Efficient intelligent factory scheduling system based on deep learning | |
CN115186935A (en) | Electromechanical device nonlinear fault prediction method and system | |
CN112464578B (en) | Amplification method for predicting degradation data of residual service life | |
CN103605355A (en) | Method and system for rapidly and remotely monitoring vibration parameters of steam turbine generator unit in online | |
CN112800140A (en) | High-reliability data acquisition method based on block chain prediction machine | |
CN112926401A (en) | Transmission line hardware corrosion detection method and system | |
CN116466183B (en) | 5G-based power transmission line fault point monitoring system | |
CN117493498B (en) | Electric power data mining and analysis system based on industrial Internet | |
CN108875205A (en) | System availability efficient simulation method based on reachable matrix and discrete event driving |
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 |