CN113807431A - Intelligent spindle state evaluation method and system based on multi-source information fusion - Google Patents

Intelligent spindle state evaluation method and system based on multi-source information fusion Download PDF

Info

Publication number
CN113807431A
CN113807431A CN202111083112.4A CN202111083112A CN113807431A CN 113807431 A CN113807431 A CN 113807431A CN 202111083112 A CN202111083112 A CN 202111083112A CN 113807431 A CN113807431 A CN 113807431A
Authority
CN
China
Prior art keywords
intelligent
fusion
module
analysis
processing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111083112.4A
Other languages
Chinese (zh)
Inventor
张燕飞
李赟豪
孔令飞
元振毅
王丽洁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian University of Technology
Original Assignee
Xian University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xian University of Technology filed Critical Xian University of Technology
Priority to CN202111083112.4A priority Critical patent/CN113807431A/en
Publication of CN113807431A publication Critical patent/CN113807431A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/251—Fusion techniques of input or preprocessed data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/253—Fusion techniques of extracted features
    • 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
    • G06N3/047—Probabilistic or stochastic 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
    • 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
    • G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

The invention discloses an intelligent spindle state evaluation method and system based on multi-source information fusion, belongs to the technical field of intelligent manufacturing, and particularly comprises an original signal acquisition and preprocessing, original signal feature extraction, multi-source information fusion processing and a probabilistic neural network diagnosis model. The original signal feature extraction mainly comprises the steps of carrying out time domain, frequency domain and time-frequency domain correlation processing on the acquired data to ensure that the extracted information features reflect the service state of a main shaft more completely; carrying out fusion processing on the characteristic information; and (3) building a probabilistic neural network diagnosis model, classifying the fused fault feature information, acquiring a diagnosis result, and finally performing state evaluation on the diagnosis result. The intelligent spindle state evaluation method provided by the invention realizes the on-line evaluation of the intelligent spindle state, has an important guiding function on the evaluation of the service state of the intelligent spindle, and simultaneously improves the economic benefit.

Description

Intelligent spindle state evaluation method and system based on multi-source information fusion
Technical Field
The invention belongs to the technical field of intelligent manufacturing, and relates to an intelligent spindle state evaluation method and system based on multi-source information fusion.
Background
Intelligent manufacturing techniques play a leading role in the contemporary global manufacturing industry. In recent years, with the continuous development of information technology, especially advanced technologies such as artificial intelligence, big data, internet of things, cloud computing and the like are widely applied in engineering practice, and the manufacturing industry revolution with the goal of intelligent manufacturing is further promoted. The modern manufacturing industry is continuously developing towards informatization, networking, intellectualization, service and greening. The development of flexibility, integration and intellectualization of the numerical control machine tool is particularly important. The main shaft is used as a core component of a machine tool, the performance level of the main shaft determines the precision of a manufactured finished product, the traditional main shaft takes a motor as a power source and is adjusted by means of mechanical transmission mechanisms such as gears, belts and chain wheels, the intelligent main shaft integrates the motor and the main shaft into a whole, electromechanical coupling transmission is achieved, complicated transmission components in the middle are reduced, and transmission efficiency is improved. The intelligent judgment standard depends on whether the equipment has the functions of autonomous sensing and decision-making while having the execution capacity, and for the intelligent spindle, the intelligent spindle can be monitored in real time through a displacement sensor, an acceleration sensor, a temperature sensor and the like, so that the state information of the spindle, such as radial deviation, abnormal vibration, excessive local temperature and the like in the operation process can be timely monitored, corresponding handling is made, the processing efficiency and the processing precision are ensured, and the processing cost is further reduced.
The main shaft vibration is caused by various reasons, such as main shaft rotor balance, main shaft local cracks, main shaft rotation errors, workpiece and tool installation errors and the like. The service state of the main shaft is evaluated, essentially, the running state of the main shaft is monitored, the vibration state of the main shaft is observed in time by processing the measured vibration signal, and the state is evaluated. The current state evaluation method can be divided into fault principle-based state evaluation, multi-sensor information fusion technology-based state evaluation, vibration signal feature extraction-based state evaluation and intelligent algorithm-based state evaluation. All methods involve the collection of original data and the processing of collected data, and the traditional fault principle evaluation is not suitable for solving the problem of intelligent spindle vibration information detection under big data; for the evaluation method of the multi-sensor information fusion technology, the pixel layer fusion method has extremely low data processing effect and consumes a long time, and a large amount of detailed information is easily lost by the decision layer fusion method.
Disclosure of Invention
The invention aims to overcome the defects that the pixel layer fusion method has extremely low data processing effect and consumes a lot of time and the decision layer fusion method is easy to lose a large amount of detailed information in the prior art, and provides an intelligent spindle state evaluation method and system based on multi-source information fusion.
In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:
an intelligent spindle state evaluation method based on multi-source information fusion comprises the following steps:
collecting an original signal of the motion of the intelligent spindle;
carrying out time domain, frequency domain and time-frequency domain correlation characteristic analysis on the acquired original signal to obtain characteristic information;
carrying out fusion processing on the characteristic information;
and establishing a probabilistic neural network diagnosis model, and classifying based on the feature information after fusion processing to obtain the diagnosis result of the intelligent spindle state.
Preferably, the correlation features, when analyzed,
the time domain analysis comprises analysis processing of a peak index, a mean index and a root mean square;
the frequency domain analysis comprises analysis processing of a local peak spectrum, a mean frequency and a mean square frequency;
the time-frequency analysis analyzes the time-frequency distribution of the vibration through short-time Fourier transform.
Preferably, the fusing processing of the characteristic information is to perform the fusing processing of the characteristic information by adopting an improved DBSCAN cluster analysis algorithm;
during fusion processing, firstly, a two-dimensional vibration characteristic data set is obtained based on characteristic information, then the two-dimensional vibration characteristic data set is used as an input value, and an obtained output value is used as a data set after fusion processing and Eps and MinPts parameters obtained through clustering analysis.
Preferably, the diagnosis process of the intelligent spindle state is as follows:
calculating the health index of the intelligent spindle according to the Eps and MinPts parameters obtained by clustering analysis based on the feature data after fusion processing, and then classifying the feature data to form a plurality of groups of sample sets;
selecting a plurality of groups of the training samples to construct a probabilistic neural network diagnosis model, taking the rest samples as test samples, checking the classification effect of the network through the back substitution of the training samples, verifying the diagnosis precision of the probabilistic neural network model, and adjusting according to whether the test result meets the expected requirement until the expected result is achieved;
and finally, diagnosing the network model of the rest test samples, and further verifying the capability of the network model for diagnosing the state of the intelligent spindle.
Preferably, the diagnostic process of the state of the intelligent spindle specifically comprises:
s1, defining health index according to Eps and MinPts parameters as follows:
Figure BDA0003264706210000031
wherein D is the minimum value of the distance from the sampling point to the core object; HI is a health index, and HI is more than or equal to 0 and less than or equal to 100;
setting a health state number corresponding to the health index;
s2, classifying the health indexes of the output data set, and selecting a plurality of groups of samples as input values of the probabilistic neural network diagnosis model;
and S3, taking all the health state numbers as output values of the probabilistic neural network diagnosis model to obtain state classification results.
Preferably, the original signal comprises a power supply signal, a sensor driving signal, a signal conditioning signal and a signal storage signal;
before the related characteristic analysis is carried out on the original signal, trend removing processing is firstly carried out;
an intelligent spindle state evaluation system based on multi-source information fusion comprises:
the signal acquisition unit is used for acquiring an original signal of the intelligent spindle motion;
the characteristic information extraction unit is interacted with the signal acquisition unit and is used for carrying out time domain, frequency domain and time-frequency domain related characteristic analysis on the acquired original signal to obtain characteristic information;
the information processing module is interacted with the characteristic information extraction unit and is used for carrying out fusion processing on the characteristic information;
the probabilistic neural network diagnosis model unit is interacted with the information processing module and used for establishing a probabilistic neural network diagnosis model;
and the evaluation and diagnosis module is respectively interacted with the probabilistic neural network diagnosis model unit and the information processing module, and performs classified evaluation by combining the feature information after fusion processing on the basis of the probabilistic neural network diagnosis model to obtain the diagnosis result of the state of the intelligent spindle.
Preferably, the signal acquisition unit comprises a power signal acquisition module, a sensor driving module, a signal conditioning module and a signal storage module, and the power signal, the sensor driving signal, the signal processing information and the signal storage information are respectively acquired.
Preferably, the sensor driving module is realized by a three-way acceleration sensor and a laser displacement sensor,
the three-way acceleration sensor is arranged at the front bearing and the rear bearing of the intelligent main shaft and used for measuring the vibration condition of the intelligent main shaft; the laser displacement sensor is arranged at the front end of the intelligent spindle and used for measuring the jumping condition of the intelligent spindle;
the signal conditioning module is realized by a filter and an amplifier, the filter is used for removing signals irrelevant to the vibration of the intelligent spindle, and the amplifier is used for amplifying the obtained signals.
Preferably, the feature information extraction unit includes a time domain analysis module, a frequency domain analysis module and a time-frequency analysis module,
the time domain analysis module is used for analyzing and processing a peak index, a mean index and a root mean square of the original signal;
the frequency domain analysis module is used for analyzing and processing a local peak spectrum, an average frequency and a mean square frequency of an original signal;
the time-frequency analysis is used to analyze the time-frequency distribution of the original signal based on a short-time fourier transform.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses an intelligent spindle state evaluation method based on multi-source information fusion, which comprises the following steps: the method comprises the steps of raw signal acquisition and preprocessing, raw signal feature extraction, multi-source information fusion processing and a probabilistic neural network diagnosis model. The original signal feature extraction mainly comprises the steps of carrying out time domain, frequency domain and time-frequency domain correlation processing on the acquired data to ensure that the extracted information features reflect the service state of a main shaft more completely; carrying out fusion processing on the characteristic information; and (3) building a probabilistic neural network diagnosis model, classifying the fused fault feature information, acquiring a diagnosis result, and finally performing state evaluation on the diagnosis result. The intelligent spindle state evaluation method provided by the invention realizes the on-line evaluation of the intelligent spindle state, has an important guiding function on the evaluation of the service state of the intelligent spindle, and simultaneously improves the economic benefit.
Furthermore, vibration signal acquisition is mainly completed by building a data acquisition platform and is composed of an intelligent spindle, a power supply part, a sensor driving part, a signal conditioning part and a signal storage part, so that original data during spindle running are accurately acquired, and a data basis is provided for intelligent spindle state evaluation.
Further, the original signal feature extraction mainly comprises the steps of extracting relevant features of a time domain, a frequency domain and a time-frequency domain of the acquired data so as to ensure that the extracted features better reflect the vibration state of the main shaft.
And further, multi-source information fusion processing is carried out, the feature information is fused by adopting an improved DBSCAN clustering analysis algorithm, the optimal Eps and MinPts parameters are obtained to define a Health Index (HI), and a basis is provided for classification of the spindle state.
Further, a Probabilistic Neural Network (PNN) is used as a mode classification method, the fusion processing result is classified through a Health Index (HI) to obtain a sample set, a plurality of groups in the sample set are selected to train the PNN network model to obtain a diagnosis model, the rest groups in the sample set are substituted into the diagnosis model to verify the recognition accuracy of the model, and finally the diagnosis result is obtained.
Drawings
FIG. 1 is a schematic diagram of a signal acquisition hardware platform according to the present invention in FIG. 1;
FIG. 2 is a flow chart of the intelligent spindle state evaluation based on multi-source information fusion according to the present invention;
FIG. 3 is a flow chart of the improved DBSCAN cluster analysis algorithm of the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
example 1
An intelligent spindle state evaluation method based on multi-source information fusion comprises the following steps:
collecting an original signal of the motion of the intelligent spindle;
carrying out time domain, frequency domain and time-frequency domain correlation characteristic analysis on the acquired original signal to obtain characteristic information;
carrying out fusion processing on the characteristic information;
and establishing a probabilistic neural network diagnosis model, and classifying based on the feature information after fusion processing to obtain the diagnosis result of the intelligent spindle state.
Example 2
As shown in fig. 1, an intelligent spindle state monitoring signal acquisition hardware platform based on multi-source information fusion includes an intelligent spindle, a dc power supply part, a sensor part, a signal conditioning part, and a signal storage part. Wherein the voltage of the direct current power supply is 24V; the sensor part comprises 3 acceleration sensors for measuring vibration states and 1 laser displacement sensor for measuring rotor displacement; the signal conditioning part comprises a high-pass filter and an amplifier and is used for reducing redundancy and removing noise; the storage part stores the processed signals into the data acquisition card to prepare for feature extraction.
As shown in fig. 2, the intelligent spindle state evaluation flowchart based on multi-source information fusion includes the following steps:
A. collecting and preprocessing original signals;
B. extracting original signal features;
C. fusing feature layer information based on an improved DBSCAN fusion algorithm;
D. a Probabilistic Neural Network (PNN) model;
E. state classification based on intelligent spindle Health Index (HI);
F. a PNN state diagnostic model;
G. whether the PNN model diagnosis result meets an expected value;
H. and outputting an evaluation result.
Specifically, in step B, the vibration feature information is extracted through a time domain, a frequency domain, and a time-frequency domain. The time domain analysis comprises a peak index, a mean index and a root mean square, the frequency domain analysis comprises a local peak spectrum, a mean frequency and a mean square frequency, and the time frequency analysis represents the vibration time-frequency distribution condition through short-time Fourier transform (STFT).
Specifically, in the step G, whether the PNN model diagnosis result meets the expected value or not, for the intelligent spindle state category divided by the defined HI index, sample data is input into the PNN model, whether the error of the classification result meets the design requirement or not is observed, and if not, the PNN model is retrained; and if so, outputting a state classification result.
As shown in fig. 3, the modified DBSCAN cluster analysis algorithm flowchart specifically includes the following steps:
calculating Euclidean distance between data points in data set to obtain distance matrix Yn×n。
② mixing Yn×nArranging the elements in each row in an ascending order, and calculating the average value of each row;
taking the average value as a candidate set Y of EpsEps；
Fourthly, calculating Yn×n≤YEpsThe number N of the MinPts is taken as a candidate parameter of the MinPts;
adopting Eps and MinPts parameters of each group to perform DBSCAN clustering (stopping calculation when the number of clustered clusters is 1);
sixthly, calculating the minimum value of the intra-cluster density and the inter-cluster density to obtain corresponding Eps and MinPts parameters;
and defining attributes: the core point is 1, the boundary point is 0, and the noise point is-1;
traversing all data points { S }i,jIf S11If the core point is not the boundary point, judging whether the core point is the boundary point, and if the core point is the boundary point, judging whether the core point is the boundary point. The index is 0 until all sample points have been traversed.
Ninthly, acquiring sample points with the labels of 0 and 1, and eliminating abnormal points with the labels of-1.
Specifically, in step sixthly, the Health Index (HI) is defined for obtaining the optimal Eps and MinPts parameters, and the calculation formula is:
Figure BDA0003264706210000081
wherein D is the minimum of sampling point to core object distance, and the healthy index HI of rotary equipment is between 0 ~ 100, among the practical engineering application: (80, 100) shows that the healthy state is very good and is numbered 1, (60, 80) shows that the healthy state is good and is numbered 2, (40, 60) shows that an abnormality occurs and is numbered 3, (20, 40) shows that the disease state is 4, and [0, 20] shows that the disease state is serious and is numbered 5.
Collecting information of the working state of the main shaft through a plurality of acceleration sensors and laser displacement sensors, and extracting vibration characteristics of the main shaft by using a time domain, frequency domain and time-frequency domain characteristic extraction method; performing fusion processing on the characteristic information by adopting an improved DBSCAN clustering analysis algorithm, and classifying the fused samples through a rotating equipment health index HI; and (3) building a PNN diagnosis model, training the fused state characteristic information to obtain a diagnosis result, and finally performing state evaluation on the diagnosis result.
In summary, the evaluation method of the present invention is an intelligent spindle state evaluation method based on multi-source information fusion. The vibration and the main shaft jumping information of the intelligent main shaft running state are collected, and all state signals are analyzed and processed, so that the method particularly relates to the use of improving a DBSCAN cluster analysis fusion algorithm and establishing a PNN classification model, and further provides a reference result for the evaluation of the main shaft running state.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.

Claims (10)

1. An intelligent spindle state evaluation method based on multi-source information fusion is characterized by comprising the following steps:
collecting an original signal of the motion of the intelligent spindle;
carrying out time domain, frequency domain and time-frequency domain correlation characteristic analysis on the acquired original signal to obtain characteristic information;
carrying out fusion processing on the characteristic information;
and establishing a probabilistic neural network diagnosis model, and classifying based on the feature information after fusion processing to obtain the diagnosis result of the intelligent spindle state.
2. The intelligent spindle state evaluation method based on multi-source information fusion of claim 1, wherein during the analysis of relevant characteristics,
the time domain analysis comprises analysis processing of a peak index, a mean index and a root mean square;
the frequency domain analysis comprises analysis processing of a local peak spectrum, a mean frequency and a mean square frequency;
the time-frequency analysis analyzes the time-frequency distribution of the vibration through short-time Fourier transform.
3. The intelligent spindle state evaluation method based on multi-source information fusion of claim 1, wherein the fusion processing of the characteristic information is the fusion processing of the characteristic information by adopting an improved DBSCAN cluster analysis algorithm;
during fusion processing, firstly, a two-dimensional vibration characteristic data set is obtained based on characteristic information, then the two-dimensional vibration characteristic data set is used as an input value, and an obtained output value is used as a data set after fusion processing and Eps and MinPts parameters obtained through clustering analysis.
4. The intelligent spindle state evaluation method based on multi-source information fusion of claim 3, wherein the diagnosis process of the intelligent spindle state is as follows:
calculating the health index of the intelligent spindle according to the Eps and MinPts parameters obtained by clustering analysis based on the feature data after fusion processing, and then classifying the feature data to form a plurality of groups of sample sets;
selecting a plurality of groups of the training samples to construct a probabilistic neural network diagnosis model, taking the rest samples as test samples, checking the classification effect of the network through the back substitution of the training samples, verifying the diagnosis precision of the probabilistic neural network model, and adjusting according to whether the test result meets the expected requirement until the expected result is achieved;
and finally, diagnosing the network model of the rest test samples, and further verifying the capability of the network model for diagnosing the state of the intelligent spindle.
5. The intelligent spindle state evaluation method based on multi-source information fusion according to claim 3, wherein the diagnosis process of the intelligent spindle state is specifically as follows:
s1, defining health index according to Eps and MinPts parameters as follows:
Figure FDA0003264706200000021
wherein D is the minimum value of the distance from the sampling point to the core object; HI is a health index, and HI is more than or equal to 0 and less than or equal to 100;
setting a health state number corresponding to the health index;
s2, classifying the health indexes of the output data set, and selecting a plurality of groups of samples as input values of the probabilistic neural network diagnosis model;
and S3, taking all the health state numbers as output values of the probabilistic neural network diagnosis model to obtain state classification results.
6. The intelligent spindle state evaluation method based on multi-source information fusion of claim 1, wherein the original signals comprise power signals, sensor driving signals, signal conditioning signals and signal storage signals;
the raw signal is first detrended before being subjected to correlation feature analysis.
7. The utility model provides an intelligence main shaft state evaluation system based on multisource information fusion which characterized in that includes:
the signal acquisition unit is used for acquiring an original signal of the intelligent spindle motion;
the characteristic information extraction unit is interacted with the signal acquisition unit and is used for carrying out time domain, frequency domain and time-frequency domain related characteristic analysis on the acquired original signal to obtain characteristic information;
the information processing module is interacted with the characteristic information extraction unit and is used for carrying out fusion processing on the characteristic information;
the probabilistic neural network diagnosis model unit is interacted with the information processing module and used for establishing a probabilistic neural network diagnosis model;
and the evaluation and diagnosis module is respectively interacted with the probabilistic neural network diagnosis model unit and the information processing module, and performs classified evaluation by combining the feature information after fusion processing on the basis of the probabilistic neural network diagnosis model to obtain the diagnosis result of the state of the intelligent spindle.
8. The multi-source information fusion-based intelligent spindle state evaluation system according to claim 7, wherein the signal acquisition unit comprises a power signal acquisition module, a sensor driving module, a signal conditioning module and a signal storage module, and the power signal acquisition module, the sensor driving module, the signal conditioning module and the signal storage module respectively acquire power signals, sensor driving signals, signal processing information and signal storage information.
9. The intelligent spindle state evaluation system based on multi-source information fusion of claim 8, wherein the sensor driving module is realized by a three-way acceleration sensor and a laser displacement sensor,
the three-way acceleration sensor is arranged at the front bearing and the rear bearing of the intelligent main shaft and used for measuring the vibration condition of the intelligent main shaft; the laser displacement sensor is arranged at the front end of the intelligent spindle and used for measuring the jumping condition of the intelligent spindle;
the signal conditioning module is realized by a filter and an amplifier, the filter is used for removing signals irrelevant to the vibration of the intelligent spindle, and the amplifier is used for amplifying the obtained signals.
10. The multi-source information fusion-based intelligent spindle state evaluation system of claim 7, wherein the feature information extraction unit comprises a time domain analysis module, a frequency domain analysis module and a time-frequency analysis module,
the time domain analysis module is used for analyzing and processing a peak index, a mean index and a root mean square of the original signal;
the frequency domain analysis module is used for analyzing and processing a local peak spectrum, an average frequency and a mean square frequency of an original signal;
the time-frequency analysis is used to analyze the time-frequency distribution of the original signal based on a short-time fourier transform.
CN202111083112.4A 2021-09-15 2021-09-15 Intelligent spindle state evaluation method and system based on multi-source information fusion Pending CN113807431A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111083112.4A CN113807431A (en) 2021-09-15 2021-09-15 Intelligent spindle state evaluation method and system based on multi-source information fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111083112.4A CN113807431A (en) 2021-09-15 2021-09-15 Intelligent spindle state evaluation method and system based on multi-source information fusion

Publications (1)

Publication Number Publication Date
CN113807431A true CN113807431A (en) 2021-12-17

Family

ID=78895459

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111083112.4A Pending CN113807431A (en) 2021-09-15 2021-09-15 Intelligent spindle state evaluation method and system based on multi-source information fusion

Country Status (1)

Country Link
CN (1) CN113807431A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114638280A (en) * 2022-01-30 2022-06-17 江苏东佳电气有限公司 Explosion-proof motor bearing temperature anomaly monitoring system based on local equipment network

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111582392A (en) * 2020-05-12 2020-08-25 河北工业大学 Multi-working-condition health state online monitoring method for key components of wind turbine generator
CN112034789A (en) * 2020-08-25 2020-12-04 国家机床质量监督检验中心 Health assessment method, system and assessment terminal for key parts and complete machine of numerical control machine tool
CN113344134A (en) * 2021-06-30 2021-09-03 广东电网有限责任公司 Data acquisition abnormity detection method and system for low-voltage power distribution monitoring terminal

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111582392A (en) * 2020-05-12 2020-08-25 河北工业大学 Multi-working-condition health state online monitoring method for key components of wind turbine generator
CN112034789A (en) * 2020-08-25 2020-12-04 国家机床质量监督检验中心 Health assessment method, system and assessment terminal for key parts and complete machine of numerical control machine tool
CN113344134A (en) * 2021-06-30 2021-09-03 广东电网有限责任公司 Data acquisition abnormity detection method and system for low-voltage power distribution monitoring terminal

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
于凯 等: "《基于参数自适应DBSCAN算法的旋转设备健康评估》", 《电工电气》, no. 12, pages 24 - 29 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114638280A (en) * 2022-01-30 2022-06-17 江苏东佳电气有限公司 Explosion-proof motor bearing temperature anomaly monitoring system based on local equipment network
CN114638280B (en) * 2022-01-30 2023-04-11 江苏东佳电气有限公司 Explosion-proof motor bearing temperature anomaly monitoring system based on local equipment network

Similar Documents

Publication Publication Date Title
CN108444708B (en) Method for establishing rolling bearing intelligent diagnosis model based on convolutional neural network
CN111914883B (en) Spindle bearing state evaluation method and device based on deep fusion network
CN109460618B (en) Rolling bearing residual life online prediction method and system
CN111623868B (en) Convolutional neural network construction method for rail corrugation identification
CN111637964B (en) Rail corrugation identification method
CN111046945B (en) Fault type and damage degree diagnosis method based on combined convolutional neural network
CN110110768B (en) Rolling bearing fault diagnosis method based on parallel feature learning and multiple classifiers
CN110070060B (en) Fault diagnosis method for bearing equipment
CN110132554B (en) Rotary machine fault diagnosis method based on deep Laplace self-coding
CN112257530B (en) Rolling bearing fault diagnosis method based on blind signal separation and support vector machine
CN111562108A (en) Rolling bearing intelligent fault diagnosis method based on CNN and FCMC
CN113188794B (en) Gearbox fault diagnosis method and device based on improved PSO-BP neural network
CN111947928A (en) Multi-source information fusion bearing fault prediction system and method
CN109946080B (en) Mechanical equipment health state identification method based on embedded circulation network
CN115409131B (en) Production line abnormity detection method based on SPC process control system
CN114755017B (en) Variable-speed bearing fault diagnosis method of cross-domain data driving unsupervised field shared network
CN116793682A (en) Bearing fault diagnosis method based on iCORAL-MMD and anti-migration learning
CN114462480A (en) Multi-source sensor rolling mill fault diagnosis method based on non-equilibrium data set
CN114429152A (en) Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption
CN113850161A (en) Flywheel fault identification method based on LSTM deep noise reduction self-encoder
CN116956215A (en) Fault diagnosis method and system for transmission system
CN115587290A (en) Aero-engine fault diagnosis method based on variational self-coding generation countermeasure network
CN113807431A (en) Intelligent spindle state evaluation method and system based on multi-source information fusion
CN114354184A (en) Deep learning-based method and device for establishing health early warning model of main shaft of large-scale rotating equipment
CN113092044B (en) Rotary machine fault diagnosis method based on weighted level visible graph

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