CN110815224A - Remote fault diagnosis pushing method and device for robot - Google Patents
Remote fault diagnosis pushing method and device for robot Download PDFInfo
- Publication number
- CN110815224A CN110815224A CN201911114796.2A CN201911114796A CN110815224A CN 110815224 A CN110815224 A CN 110815224A CN 201911114796 A CN201911114796 A CN 201911114796A CN 110815224 A CN110815224 A CN 110815224A
- Authority
- CN
- China
- Prior art keywords
- robot
- fault
- signal
- state
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1679—Programme controls characterised by the tasks executed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1679—Programme controls characterised by the tasks executed
- B25J9/1689—Teleoperation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
Abstract
The invention discloses a remote fault diagnosis pushing method and device for a robot, wherein the method comprises the following steps: acquiring a robot state signal and a working environment state signal, preprocessing the signals and acquiring preprocessed signals; performing characteristic extraction on the signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal; fusing state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristics of the robot; and substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault judgment model library, judging that the robot has a fault through logic calculation, acquiring the fault type of the robot, and pushing the fault type to a management user terminal for real-time display of fault alarm. In the implementation of the invention, whether the robot has a fault can be remotely diagnosed in real time and pushed to a corresponding manager, so that the fault diagnosis efficiency of the robot is improved.
Description
Technical Field
The invention relates to the technical field of remote fault monitoring of robots, in particular to a remote fault diagnosis pushing method and device of a robot.
Background
With the continuous improvement of the use density of the robot, an automatic, intelligent and flexible production line formed by the robot is increasingly applied to various industrial fields such as welding, spraying, assembling, packaging, carrying, stacking and the like. The safe, stable, continuous and reliable operation of the robot can bring considerable economic benefits to enterprises, and once the robot is stopped for a long time due to failure, the inevitable economic loss can be caused to the enterprises.
The large number of robots of different brands used in various complex production environments presents a significant challenge to the routine maintenance of the robots. Common failure modes of robots include: mechanical system faults, electrical system faults, sensing detection system faults, control system faults, and the like. The operation and maintenance mode of the robot depending on single-machine fault alarm or manual inspection has the defects of high maintenance cost, long response time and the like, and the current actual needs of enterprises can not be met more and more.
At present, ABB and FANUC companies in four major families of foreign robots have developed respective robot remote operation and maintenance systems. A remote service platform developed by ABB can carry out remote real-time monitoring and data analysis processing on robots of client enterprises, faults possibly occurring are pre-judged at the first time, an alarm mechanism is started, and a service engineer logs in through the Internet to receive fault information and provides technical support for clients. The FANUC company also provides a self remote monitoring service scheme, finds out possible faults by detecting the change of the state parameters of the robot, and timely informs field operators and service engineers.
However, these robot remote maintenance systems are generally only for corresponding robots produced by themselves, but are not compatible with robots produced by other manufacturers, and in terms of the use of the robots, there is often a situation that multiple robots of multiple manufacturers are used together at the same time, so that corresponding fault remote monitoring cannot be performed in time, and thus the maintenance efficiency is low.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a remote fault diagnosis pushing method and device of a robot, which are used for improving the fault diagnosis efficiency of the robot.
In order to solve the technical problem, an embodiment of the present invention further provides a remote fault diagnosis pushing method for a robot, where the method includes:
acquiring a robot state signal and a working environment state signal based on each sensor;
preprocessing the collected robot state signals and the collected working environment state signals to obtain preprocessed signals;
performing characteristic extraction on the signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal;
fusing the state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristic of the robot;
substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault judgment model library, and judging whether the robot has a fault at present through logic calculation;
and if so, acquiring the fault type of the robot, and pushing the fault type to a management user terminal for real-time display of fault alarm.
Optionally, the preprocessing the collected robot state signal and the collected work environment state signal to obtain a preprocessed signal includes:
sequentially carrying out analog/digital conversion, coding and denoising on the collected robot state signal and the collected operation environment state signal to obtain a processed state signal;
and carrying out data cleaning processing on the processed state signal to obtain a preprocessed signal.
Optionally, the performing analog-to-digital conversion, encoding and denoising on the collected robot state signal and the collected work environment state signal in sequence includes:
the collected robot state signals and the collected working environment state signals are sequentially subjected to sampling, holding and quantification processing, and processing results are obtained; and the number of the first and second groups,
carrying out unique identification coding on the processing result according to a time sequence and signal type mode;
and denoising the data subjected to the uniqueness identification coding based on digital filtering.
Optionally, the performing data cleaning processing on the processed status signal includes:
deleting invalidity and repeated data processing are carried out on the processed state signals based on a sorting and merging algorithm;
judging abnormal values of the state signals after invalid deletion and repeated data processing based on a three-sigma criterion, and carrying out abnormal data correction processing to obtain corrected data;
and performing complement missing data processing on the corrected data based on a difference method.
Optionally, the feature extraction of the signal based on the time domain analysis and the frequency domain analysis includes:
performing feature extraction on the cleaned state signal by adopting a signal statistical analysis algorithm based on time domain analysis and frequency domain analysis;
wherein, the time domain analysis comprises extracting the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value of the signal; the frequency domain analysis includes extracting a frequency response peak, a frequency spectrum, a power density spectrum of the signal.
Optionally, the fusing the state feature data of the signal based on the principal analysis algorithm to obtain the running state feature of the robot includes:
respectively carrying out multi-scale decomposition on the state characteristic data of the signal to be fused to obtain a low-pass sub-band component and a band-pass sub-band component of the state characteristic data of the signal to be fused;
fusing the low-pass sub-band components in the state feature data of the signal to be fused based on a first fusion rule of a principal component analysis algorithm to obtain a first fusion feature;
fusing the band-pass sub-band components in the state feature data of the signal to be fused based on a second fusion rule of the principal component analysis algorithm to obtain a second fusion feature;
performing inverse transformation fusion on the first fusion characteristic and the second fusion characteristic to obtain the running state characteristic of the robot;
the first fusion rule is a fusion rule of a principal component analysis algorithm and non-negative matrix factorization; the second fusion rule is a fusion rule of a principal component analysis algorithm and a neural network model.
Optionally, substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault discrimination model library, and determining whether the robot has a fault currently through logic calculation includes:
and respectively substituting the running state characteristics of the robot into a multi-state fault tree model of the whole robot and a multi-element logistic regression model of the key parts, which are stored in a fault discrimination model library, and judging whether the current whole robot or the key parts of the robot have faults or not through logical calculation.
Optionally, the obtaining the fault type of the robot and pushing the fault type to a management user terminal for real-time displaying of a fault alarm includes:
and acquiring the fault type of the robot as a complete machine fault or a key part fault, and pushing the fault type to a corresponding management user terminal to display a fault alarm in real time based on an HTTP (hyper text transport protocol).
Optionally, the method further includes:
matching in a maintenance knowledge base according to the fault types of the robot to obtain similar fault types matched with each other;
performing maintenance information retrieval in the maintenance knowledge base by using the mutually matched similar fault types, and sequencing retrieval results to obtain the sequenced maintenance information of the mutually matched similar fault types;
and pushing the sequenced maintenance information of the similar fault types matched with each other to a corresponding management user terminal based on an HTTP transmission protocol for maintenance suggestion display.
In addition, the embodiment of the invention also provides a remote fault diagnosis pushing device of the robot, which comprises:
a data acquisition module: the robot state signal acquisition device is used for acquiring a robot state signal and a working environment state signal based on each sensor;
a data preprocessing module: the system comprises a robot state signal acquisition unit, a robot state signal acquisition unit and a working environment state signal acquisition unit, wherein the robot state signal acquisition unit is used for acquiring a robot state signal and a working environment state signal;
a feature extraction module: the system comprises a signal processing module, a signal analyzing module and a signal analyzing module, wherein the signal processing module is used for carrying out characteristic extraction on a signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal;
a feature fusion module: the system is used for fusing state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristics of the robot;
and a fault judging module: the fault diagnosis module is used for substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault judgment model library and judging whether the robot has a fault at present through logic calculation;
failure early warning propelling movement module: and the fault type display module is used for acquiring the fault type of the robot and pushing the fault type to a management user terminal for real-time display of fault alarm if the fault is judged to occur.
In the embodiment of the invention, whether the robot has a fault can be remotely diagnosed in real time and pushed to a corresponding manager, so that the fault diagnosis efficiency of the robot is improved; the remote fault diagnosis and monitoring of robots of different brands and models of different manufacturers can be compatible; the whole monitoring and calculating process is finished at the cloud end, so that the calculating capacity is improved, and the response speed is increased; the robot fault diagnosis method based on the polymorphic fault tree model and the logistic regression model is provided, so that fault diagnosis becomes more accurate and efficient.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a remote fault diagnosis pushing method of a robot in an embodiment of the present invention;
fig. 2 is a schematic structural composition diagram of a remote fault diagnosis pushing device of a robot in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
Referring to fig. 1, fig. 1 is a schematic flow chart of a remote fault diagnosis pushing method for a robot according to an embodiment of the present invention.
As shown in fig. 1, a remote fault diagnosis pushing method for a robot includes:
s11: acquiring a robot state signal and a working environment state signal based on each sensor;
in the specific implementation process of the invention, the robot state signals comprise current state signals, corner signals, position signals, speed signals, angular speed signals, acceleration signals, vibration signals and shaft temperature signals; the work environment status signal comprises: an ambient temperature signal, an ambient humidity signal, and an ambient noise signal.
Specifically, different sensors are arranged on different key components and surrounding environments of the robot, for example, a sensor for collecting current, a sensor for collecting angular velocity, and the like are arranged on the robot; a sensor for collecting the ambient temperature, a sensor for collecting the ambient humidity and the like are arranged at the periphery of the robot; the acquisition of the robot status signals by the different sensors comprises: current status signals, corner signals, position signals, speed signals, angular velocity signals, acceleration signals, vibration signals, shaft temperature signals, and the like; acquiring the working environment state signal comprises: an ambient temperature signal, an ambient humidity signal, an ambient noise signal, and the like.
S12: preprocessing the collected robot state signals and the collected working environment state signals to obtain preprocessed signals;
in the specific implementation process of the present invention, the preprocessing the collected robot state signal and the operation environment state signal to obtain a preprocessed signal includes: sequentially carrying out analog/digital conversion, coding and denoising on the collected robot state signal and the collected operation environment state signal to obtain a processed state signal; and carrying out data cleaning processing on the processed state signal to obtain a preprocessed signal.
Further, the processing of analog/digital conversion, coding and denoising is performed on the collected robot state signal and the collected working environment state signal in sequence, including: the collected robot state signals and the collected working environment state signals are sequentially subjected to sampling, holding and quantification processing, and processing results are obtained; and carrying out unique identification coding on the processing result according to a time sequence and signal type mode; and denoising the data subjected to the uniqueness identification coding based on digital filtering.
Further, the performing data cleaning processing on the processed status signal includes: deleting invalidity and repeated data processing are carried out on the processed state signals based on a sorting and merging algorithm; judging abnormal values of the state signals after invalid deletion and repeated data processing based on a three-sigma criterion, and carrying out abnormal data correction processing to obtain corrected data; and performing complement missing data processing on the corrected data based on a difference method.
Specifically, each sensor collects robot state signals and working environment state signals as multi-source and heterogeneous analog signals; for convenience of subsequent processing, the signals need to be converted, encoded and denoised, wherein analog signals need to be converted into digital signals, so that subsequent reading is facilitated; coding can increase the identifiability of signals; and the noise is removed, so that the influence of impurities can be effectively reduced, and the monitoring precision is improved.
The analog/digital conversion of the collected robot status signal and the collected work environment status signal can be realized by an analog/digital converter, which specifically includes: the collected robot state signals and the collected working environment state signals are sequentially sampled, held, quantized and encoded through an analog/digital converter; the unique identification coding is carried out according to the time sequence and the type of the data signal, so that the unique data can be conveniently inquired through the coding; in the subsequent denoising, a digital filtering algorithm is adopted to filter and denoise the data signal, wherein the digital filtering comprises classical filtering and modern filtering; the classical filtering is an engineering concept proposed according to Fourier analysis and transformation, and according to high mathematics theory, any signal meeting a certain condition can be regarded as being formed by superposing infinite sine waves, namely the engineering signal is formed by linearly superposing sine waves with different frequencies, and the sine waves with different frequencies forming the signal are called frequency components or harmonic components of the signal; modern filtering is to use the nature of randomness of signals, to regard signals and their noise as random signals, and estimate the signals themselves by using their statistical characteristics, and once the signals are estimated, the obtained signals themselves are much higher than the original signal-to-noise ratio, and typical digital filters are Kalman filtering, Wenner filtering, adaptive filtering, wavelet transform (wavelet), etc.
The digital filtering has the advantages of high precision, high reliability, programmable change of characteristics or multiplexing, convenience in integration and the like; the digital filtering is of low-pass, high-pass, band-stop, all-pass and other types; may be time-invariant or time-variant, causal or non-causal, linear or non-linear; the most widely used is the linear, time-invariant digital filter.
Specifically, the processed state signals need to be screened, checked and verified, invalid and repeated data are deleted, abnormal data are corrected, and missing data are supplemented, so that the consistency and reliability of the data are ensured.
The invalid and repeated data are deleted by adopting an ordering and merging algorithm, particularly, a merging and ordering algorithm is adopted, the merging and ordering idea is simple, and the merging and ordering idea is divided into two steps and a treatment part in total; when merging and sorting, firstly, dividing an array to be sorted into two parts, namely Left and Right, and if the two parts are sorted, namely, the child arrays Left and Right are both ordered arrays; combining the two arrays, wherein the combining mode is that firstly, an array with the same capacity as the original array is created to store data during combination, then the arrays in Left and Right are compared, if Left [0] < Right [0], the Left [0] is put into the 0 index of the new array, then Left [1] and Right [0] are compared, and all the arrays in Left and Right can be copied into the new array according to a certain sequence by analogy of ascending or descending, and at the moment, the array sorting is finished; merging and sorting an array is subjected to limited-time division, then the array is subjected to merging with the same times, and the whole array is sorted.
Judging an abnormal value by adopting a three-sigma criterion to realize correction of abnormal data, wherein the abnormal value is a value of which the deviation from the average value mu exceeds 3 times of standard deviation sigma in a group of measurement values; the three sigma criterion is also called Lauda criterion, which is that firstly, a group of detection data is supposed to only contain random errors, the detection data is calculated to obtain standard deviation, an interval is determined according to a certain probability, the error exceeding the interval is considered not to belong to the random errors but to be coarse errors, and the data containing the errors are removed; the three sigma criterion is: the probability of the numerical distribution in (μ - σ, μ + σ) is 0.6827; the probability of the numerical distribution in (μ -2 σ, μ +2 σ) is 0.9545; the probability of the numerical distribution in (μ -3 σ, μ +3 σ) is 0.9973; it is considered that the values of Y are almost entirely concentrated in the (μ -3 σ, μ +3 σ) range, and the possibility of exceeding this range is only less than 0.3%.
An interpolation algorithm is adopted to realize the completion of missing data; the interpolation algorithm is also called as "interpolation method", which is a method that uses the function values of a plurality of points known in a certain interval of the function f (x) to make a proper specific function, and uses the values of the specific function as approximate values of the function f (x) at other points in the interval, and the method is called as interpolation method; specific examples thereof include Lagrange interpolation, Newton interpolation, Hermite interpolation, piecewise interpolation, spline interpolation, and the like.
S13: performing characteristic extraction on the signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal;
in a specific implementation process of the present invention, the extracting features of the signal based on the time domain analysis and the frequency domain analysis includes: performing feature extraction on the cleaned state signal by adopting a signal statistical analysis algorithm based on time domain analysis and frequency domain analysis; wherein, the time domain analysis comprises extracting the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value of the signal; the frequency domain analysis includes extracting a frequency response peak, a frequency spectrum, a power density spectrum of the signal.
Specifically, the feature extraction of the cleaned state signal is carried out in time domain analysis and frequency domain analysis; the extraction is realized through a signal statistical analysis algorithm, the extracted cleaned state signal is subjected to feature extraction, and the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value of the extracted time domain signal are obtained; frequency response peak value, frequency spectrum, power spectrum and power density spectrum of the frequency domain signal; and counting the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value in time domain analysis through a signal statistical analysis algorithm, and counting the frequency response peak value, the frequency spectrum, the power spectrum and the power density spectrum in frequency domain analysis.
S14: fusing the state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristic of the robot;
in a specific implementation process of the present invention, the fusing the state feature data of the signal based on the principal component analysis algorithm to obtain the running state feature of the robot includes: respectively carrying out multi-scale decomposition on the state characteristic data of the signal to be fused to obtain a low-pass sub-band component and a band-pass sub-band component of the state characteristic data of the signal to be fused; fusing the low-pass sub-band components in the state feature data of the signal to be fused based on a first fusion rule of a principal component analysis algorithm to obtain a first fusion feature; fusing the band-pass sub-band components in the state feature data of the signal to be fused based on a second fusion rule of the principal component analysis algorithm to obtain a second fusion feature; performing inverse transformation fusion on the first fusion characteristic and the second fusion characteristic to obtain the running state characteristic of the robot; the first fusion rule is a fusion rule of a principal component analysis algorithm and non-negative matrix factorization; the second fusion rule is a fusion rule of a principal component analysis algorithm and a neural network model.
Specifically, firstly, performing multi-scale decomposition on state characteristic data of a signal to be fused, wherein non-downsampling redundancy lifting inseparable wavelets are adopted for performing multi-scale decomposition, so that a low-pass sub-band component and a band-pass sub-band component of the state characteristic data of the signal to be fused are obtained after direction localization is completed; fusing the low-pass sub-band components in the state feature data of the signal to be fused according to a first fusion rule of a principal component analysis algorithm, so as to obtain a first fusion feature; fusing the band-pass sub-band components in the state feature data of the signal to be fused according to a second fusion rule of the principal component analysis algorithm to obtain a second fusion feature; performing inverse transformation fusion on the first fusion characteristic and the second fusion characteristic to obtain the running state characteristic of the robot; the first fusion rule is a fusion rule of a principal component analysis algorithm and non-negative matrix factorization; the second fusion rule is a fusion rule of a principal component analysis algorithm and a neural network model.
The principal component analysis algorithm comprises the following specific steps: converting the state characteristic data of the signals to be fused into a bit column vector; performing principal component analysis algorithm decomposition on the column vector to obtain a low frequency component and a sparse matrix component; the low frequency components and the sparse matrix components are converted into an m × n matrix to obtain matrix characterization data.
S15: substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault discrimination model library;
in a specific implementation process of the present invention, substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault discrimination model library includes: and respectively substituting the running state characteristics of the robot into a multi-state fault tree model of the whole robot and a multiple logistic regression model of key parts, which are stored in a fault discrimination model library.
Specifically, the fault diagnosis model comprises a polymorphic fault tree model of the whole machine and a multiple logistic regression model of the key parts, and both the polymorphic fault tree model of the whole machine and the multiple logistic regression model of the key parts are models which have been trained and converged, are stored in a fault discrimination model library, and can be directly called.
S16: judging whether the robot has a fault at present through logic calculation;
in the specific implementation process of the invention, whether the current complete machine or the key parts of the robot have faults or not is judged through the polymorphic fault tree model of the complete machine and the multiple logistic regression model of the key parts.
S17: and if so, acquiring the fault type of the robot, and pushing the fault type to a management user terminal for real-time display of fault alarm.
In a specific implementation process of the present invention, the acquiring a fault type of the robot and pushing the fault type to a management user terminal for real-time display of a fault alarm includes: and acquiring the fault type of the robot as a complete machine fault or a key part fault, and pushing the fault type to a corresponding management user terminal to display a fault alarm in real time based on an HTTP (hyper text transport protocol).
Specifically, by judging the multi-state fault tree model of the whole machine and the multiple logistic regression model of the key parts, the corresponding fault type can be obtained as the whole machine fault or the key part fault, and then the fault type is pushed to the corresponding management user terminal through the HTTP transmission protocol to carry out real-time display of fault alarm; before pushing, the Web visual module is required to process the Web visual data, and then the Web visual data is pushed to a corresponding management user terminal to display a Web visual interface so as to send out a fault alarm.
In the specific implementation process of the invention, the method further comprises the following steps: matching in a maintenance knowledge base according to the fault types of the robot to obtain similar fault types matched with each other; performing maintenance information retrieval in the maintenance knowledge base by using the mutually matched similar fault types, and sequencing retrieval results to obtain the sequenced maintenance information of the mutually matched similar fault types; and pushing the sequenced maintenance information of the similar fault types matched with each other to a corresponding management user terminal based on an HTTP transmission protocol for maintenance suggestion display.
Specifically, all maintenance schemes related to the complete machine fault and the key part fault are stored in the maintenance knowledge base, and the schemes are named by fault types, so that corresponding retrieval can be conveniently carried out through the fault types in the follow-up process; firstly, matching similar fault types in a maintenance knowledge base through the fault types; retrieving and calling corresponding maintenance information through the similar fault types, then sequencing the association degrees, and pushing part of the maintenance information with higher association degree sequencing to a corresponding management user terminal based on an HTTP (hyper text transport protocol) for maintenance suggestion display.
In the embodiment of the invention, whether the robot has a fault can be remotely diagnosed in real time and pushed to a corresponding manager, so that the fault diagnosis efficiency of the robot is improved; the remote fault diagnosis and monitoring of robots of different brands and models of different manufacturers can be compatible; the whole monitoring and calculating process is finished at the cloud end, so that the calculating capacity is improved, and the response speed is increased; the robot fault diagnosis method based on the polymorphic fault tree model and the logistic regression model is provided, so that fault diagnosis becomes more accurate and efficient.
Examples
Referring to fig. 2, fig. 2 is a schematic structural composition diagram of a remote fault diagnosis pushing device of a robot in an embodiment of the present invention.
A remote troubleshooting pushing device of a robot, the device comprising:
the data acquisition module 11: the robot state signal acquisition device is used for acquiring a robot state signal and a working environment state signal based on each sensor;
in the specific implementation process of the invention, the robot state signals comprise current state signals, corner signals, position signals, speed signals, angular speed signals, acceleration signals, vibration signals and shaft temperature signals; the work environment status signal comprises: an ambient temperature signal, an ambient humidity signal, and an ambient noise signal.
Specifically, different sensors are arranged on different key components and surrounding environments of the robot, for example, a sensor for collecting current, a sensor for collecting angular velocity, and the like are arranged on the robot; a sensor for collecting the ambient temperature, a sensor for collecting the ambient humidity and the like are arranged at the periphery of the robot; the acquisition of the robot status signals by the different sensors comprises: current status signals, corner signals, position signals, speed signals, angular velocity signals, acceleration signals, vibration signals, shaft temperature signals, and the like; acquiring the working environment state signal comprises: an ambient temperature signal, an ambient humidity signal, an ambient noise signal, and the like.
The data preprocessing module 12: the system comprises a robot state signal acquisition unit, a robot state signal acquisition unit and a working environment state signal acquisition unit, wherein the robot state signal acquisition unit is used for acquiring a robot state signal and a working environment state signal;
in the specific implementation process of the present invention, the preprocessing the collected robot state signal and the operation environment state signal to obtain a preprocessed signal includes: sequentially carrying out analog/digital conversion, coding and denoising on the collected robot state signal and the collected operation environment state signal to obtain a processed state signal; and carrying out data cleaning processing on the processed state signal to obtain a preprocessed signal.
Further, the processing of analog/digital conversion, coding and denoising is performed on the collected robot state signal and the collected working environment state signal in sequence, including: the collected robot state signals and the collected working environment state signals are sequentially subjected to sampling, holding and quantification processing, and processing results are obtained; and carrying out unique identification coding on the processing result according to a time sequence and signal type mode; and denoising the data subjected to the uniqueness identification coding based on digital filtering.
Further, the performing data cleaning processing on the processed status signal includes: deleting invalidity and repeated data processing are carried out on the processed state signals based on a sorting and merging algorithm; judging abnormal values of the state signals after invalid deletion and repeated data processing based on a three-sigma criterion, and carrying out abnormal data correction processing to obtain corrected data; and performing complement missing data processing on the corrected data based on a difference method.
Specifically, each sensor collects robot state signals and working environment state signals as multi-source and heterogeneous analog signals; for convenience of subsequent processing, the signals need to be converted and denoised, wherein analog signals need to be converted into digital signals, so that subsequent reading is facilitated; and the noise is removed, so that the influence of impurities can be effectively reduced, and the monitoring precision is improved.
The analog/digital conversion of the collected robot status signal and the collected work environment status signal can be realized by an analog/digital converter, which specifically includes: the collected robot state signals and the collected working environment state signals are sequentially sampled, held, quantized and encoded through an analog/digital converter; the unique identification coding is carried out according to the time sequence and the type of the data signal, so that the unique data can be conveniently inquired through the coding; in the subsequent denoising, a digital filtering algorithm is adopted to filter and denoise the data signal, wherein the digital filtering comprises classical filtering and modern filtering; the classical filtering is an engineering concept proposed according to Fourier analysis and transformation, and according to high mathematics theory, any signal meeting a certain condition can be regarded as being formed by superposing infinite sine waves, namely the engineering signal is formed by linearly superposing sine waves with different frequencies, and the sine waves with different frequencies forming the signal are called frequency components or harmonic components of the signal; modern filtering is to use the nature of randomness of signals, to regard signals and their noise as random signals, and estimate the signals themselves by using their statistical characteristics, and once the signals are estimated, the obtained signals themselves are much higher than the original signal-to-noise ratio, and typical digital filters are Kalman filtering, Wenner filtering, adaptive filtering, wavelet transform (wavelet), etc.
The digital filtering has the advantages of high precision, high reliability, programmable change of characteristics or multiplexing, convenience in integration and the like; the digital filtering is of low-pass, high-pass, band-stop, all-pass and other types; may be time-invariant or time-variant, causal or non-causal, linear or non-linear; the most widely used is the linear, time-invariant digital filter.
Specifically, the processed state signals need to be screened, checked and verified, invalid and repeated data are deleted, abnormal data are corrected, and missing data are supplemented, so that the consistency and reliability of the data are ensured.
The invalid and repeated data are deleted by adopting an ordering and merging algorithm, particularly, a merging and ordering algorithm is adopted, the merging and ordering idea is simple, and the merging and ordering idea is divided into two steps and a treatment part in total; when merging and sorting, firstly, dividing an array to be sorted into two parts, namely Left and Right, and if the two parts are sorted, namely, the child arrays Left and Right are both ordered arrays; combining the two arrays, wherein the combining mode is that firstly, an array with the same capacity as the original array is created to store data during combination, then the arrays in Left and Right are compared, if Left [0] < Right [0], the Left [0] is put into the 0 index of the new array, then Left [1] and Right [0] are compared, and all the arrays in Left and Right can be copied into the new array according to a certain sequence by analogy of ascending or descending, and at the moment, the array sorting is finished; merging and sorting an array is subjected to limited-time division, then the array is subjected to merging with the same times, and the whole array is sorted.
Judging an abnormal value by adopting a three-sigma criterion to realize correction of abnormal data, wherein the abnormal value is a value of which the deviation from the average value exceeds 3 times of standard deviation in a group of measurement values; the three sigma criterion is also called Lauda criterion, which is that firstly, a group of detection data is supposed to only contain random errors, the detection data is calculated to obtain standard deviation, an interval is determined according to a certain probability, the error exceeding the interval is considered not to belong to the random errors but to be coarse errors, and the data containing the errors are removed; the three sigma criterion is: the probability of the numerical distribution in (μ - σ, μ + σ) is 0.6827; the probability of the numerical distribution in (μ -2 σ, μ +2 σ) is 0.9545; the probability of the numerical distribution in (μ -3 σ, μ +3 σ) is 0.9973; it is considered that the values of Y are almost entirely concentrated in the (μ -3 σ, μ +3 σ) range, and the possibility of exceeding this range is only less than 0.3%.
An interpolation algorithm is adopted to realize the completion of missing data; the interpolation algorithm is also called as "interpolation method", which is a method that uses the function values of a plurality of points known in a certain interval of the function f (x) to make a proper specific function, and uses the values of the specific function as approximate values of the function f (x) at other points in the interval, and the method is called as interpolation method; specific examples thereof include Lagrange interpolation, Newton interpolation, Hermite interpolation, piecewise interpolation, spline interpolation, and the like.
The feature extraction module 13: the system comprises a signal processing module, a signal analyzing module and a signal analyzing module, wherein the signal processing module is used for carrying out characteristic extraction on a signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal;
in a specific implementation process of the present invention, the extracting features of the signal based on the time domain analysis and the frequency domain analysis includes: performing feature extraction on the cleaned state signal by adopting a signal statistical analysis algorithm based on time domain analysis and frequency domain analysis; wherein the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value of the time domain analysis; the frequency response peak value, the frequency spectrum, the power spectrum and the power density spectrum of the frequency domain analysis.
Specifically, the feature extraction of the cleaned state signal is carried out in time domain analysis and frequency domain analysis; the extraction is realized through a signal statistical analysis algorithm, the extracted cleaned state signal is subjected to feature extraction, and the extracted minimum value, maximum value, average value, sliding average value, variance value and root mean square value of time domain analysis are extracted; frequency response peak value, frequency spectrum, power spectrum and power density spectrum of frequency domain analysis; and counting the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value in time domain analysis through a signal statistical analysis algorithm, and counting the frequency response peak value, the frequency spectrum, the power spectrum and the power density spectrum in frequency domain analysis.
The feature fusion module 14: the system is used for fusing state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristics of the robot;
in a specific implementation process of the present invention, the fusing the state feature data of the signal based on the principal component analysis algorithm to obtain the running state feature of the robot includes: respectively carrying out multi-scale decomposition on the state characteristic data of the signal to be fused to obtain a low-pass sub-band component and a band-pass sub-band component of the state characteristic data of the signal to be fused; fusing the low-pass sub-band components in the state feature data of the signal to be fused based on a first fusion rule of a principal component analysis algorithm to obtain a first fusion feature; fusing the band-pass sub-band components in the state feature data of the signal to be fused based on a second fusion rule of the principal component analysis algorithm to obtain a second fusion feature; performing inverse transformation fusion on the first fusion characteristic and the second fusion characteristic to obtain the running state characteristic of the robot; the first fusion rule is a fusion rule of a principal component analysis algorithm and non-negative matrix factorization; the second fusion rule is a fusion rule of a principal component analysis algorithm and a neural network model.
Specifically, firstly, performing multi-scale decomposition on state characteristic data of a signal to be fused, wherein non-downsampling redundancy lifting inseparable wavelets are adopted for performing multi-scale decomposition, so that a low-pass sub-band component and a band-pass sub-band component of the state characteristic data of the signal to be fused are obtained after direction localization is completed; fusing the low-pass sub-band components in the state feature data of the signal to be fused according to a first fusion rule of a principal component analysis algorithm, so as to obtain a first fusion feature; fusing the band-pass sub-band components in the state feature data of the signal to be fused according to a second fusion rule of the principal component analysis algorithm to obtain a second fusion feature; performing inverse transformation fusion on the first fusion characteristic and the second fusion characteristic to obtain the running state characteristic of the robot; the first fusion rule is a fusion rule of a principal component analysis algorithm and non-negative matrix factorization; the second fusion rule is a fusion rule of a principal component analysis algorithm and a neural network model.
The principal component analysis algorithm comprises the following specific steps: converting the state characteristic data of the signals to be fused into a bit column vector; performing principal component analysis algorithm decomposition on the column vector to obtain a low frequency component and a sparse matrix component; the low frequency components and the sparse matrix components are converted into an m × n matrix to obtain matrix characterization data.
The failure determination module 15: the fault diagnosis module is used for substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault judgment model library and judging whether the robot has a fault at present through logic calculation;
in a specific implementation process of the present invention, substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault discrimination model library includes: and respectively substituting the running state characteristics of the robot into a multi-state fault tree model of the whole robot and a multiple logistic regression model of key parts, which are stored in a fault discrimination model library.
Specifically, the fault diagnosis model comprises a polymorphic fault tree model of the whole machine and a multiple logistic regression model of the key parts, and both the polymorphic fault tree model of the whole machine and the multiple logistic regression model of the key parts are models which have been trained and converged, are stored in a fault discrimination model library, and can be directly called.
In the specific implementation process of the invention, whether the current complete machine or the key parts of the robot have faults or not is judged through the polymorphic fault tree model of the complete machine and the multiple logistic regression model of the key parts.
The fault early warning pushing module 16: and the fault type display module is used for acquiring the fault type of the robot and pushing the fault type to a management user terminal for real-time display of fault alarm if the fault is judged to occur.
In a specific implementation process of the present invention, the acquiring a fault type of the robot and pushing the fault type to a management user terminal for real-time display of a fault alarm includes: and acquiring the fault type of the robot as a complete machine fault or a key part fault, and pushing the fault type to a corresponding management user terminal to display a fault alarm in real time based on an HTTP (hyper text transport protocol).
Specifically, by judging the multi-state fault tree model of the whole machine and the multiple logistic regression model of the key parts, the corresponding fault type can be obtained as the whole machine fault or the key part fault, and then the fault type is pushed to the corresponding management user terminal through the HTTP transmission protocol to carry out real-time display of fault alarm; before pushing, the Web visual module is required to process the Web visual data, and then the Web visual data is pushed to a corresponding management user terminal to display a Web visual interface so as to send out a fault alarm.
In the specific implementation process of the invention, the method further comprises the following steps: matching in a maintenance knowledge base according to the fault types of the robot to obtain similar fault types matched with each other; performing maintenance information retrieval in the maintenance knowledge base by using the mutually matched similar fault types, and sequencing retrieval results to obtain the sequenced maintenance information of the mutually matched similar fault types; and pushing the sequenced maintenance information of the similar fault types matched with each other to a corresponding management user terminal based on an HTTP transmission protocol for maintenance suggestion display.
Specifically, all maintenance schemes related to the complete machine fault and the key part fault are stored in the maintenance knowledge base, and the schemes are named by fault types, so that corresponding retrieval can be conveniently carried out through the fault types in the follow-up process; firstly, matching similar fault types in a maintenance knowledge base through the fault types; retrieving and calling corresponding maintenance information through the similar fault types, then sequencing the association degrees, and pushing part of the maintenance information with higher association degree sequencing to a corresponding management user terminal based on an HTTP (hyper text transport protocol) for maintenance suggestion display.
In the embodiment of the invention, whether the robot has a fault can be remotely diagnosed in real time and pushed to a corresponding manager, so that the fault diagnosis efficiency of the robot is improved; the remote fault diagnosis and monitoring of robots of different brands and models of different manufacturers can be compatible; the whole monitoring and calculating process is finished at the cloud end, so that the calculating capacity is improved, and the response speed is increased; the robot fault diagnosis method based on the polymorphic fault tree model and the logistic regression model is provided, so that fault diagnosis becomes more accurate and efficient.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic or optical disk, or the like.
In addition, the above detailed description is provided for the remote fault diagnosis pushing method and device of the robot according to the embodiment of the present invention, and a specific example is used herein to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A remote fault diagnosis pushing method for a robot is characterized by comprising the following steps:
acquiring a robot state signal and a working environment state signal based on each sensor;
preprocessing the collected robot state signals and the collected working environment state signals to obtain preprocessed signals;
performing characteristic extraction on the signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal;
fusing the state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristic of the robot;
substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault judgment model library, and judging whether the robot has a fault at present through logic calculation;
and if so, acquiring the fault type of the robot, and pushing the fault type to a management user terminal for real-time display of fault alarm.
2. The remote fault diagnosis pushing method according to claim 1, wherein the preprocessing the collected robot state signal and the work environment state signal to obtain a preprocessed signal comprises:
sequentially carrying out analog/digital conversion, coding and denoising on the collected robot state signal and the collected operation environment state signal to obtain a processed state signal;
and carrying out data cleaning processing on the processed state signal to obtain a preprocessed signal.
3. The remote fault diagnosis pushing method according to claim 2, wherein the analog/digital conversion, encoding and denoising processing are sequentially performed on the collected robot state signal and the working environment state signal, and the method comprises:
the collected robot state signals and the collected working environment state signals are sequentially subjected to sampling, holding and quantification processing, and processing results are obtained; and the number of the first and second groups,
carrying out unique identification coding on the processing result according to a time sequence and signal type mode;
and denoising the data subjected to the uniqueness identification coding based on digital filtering.
4. The remote fault diagnosis pushing method according to claim 2, wherein the performing of the data cleaning process on the processed status signal includes:
deleting invalidity and repeated data processing are carried out on the processed state signals based on a sorting and merging algorithm;
judging abnormal values of the state signals after invalid deletion and repeated data processing based on a three-sigma criterion, and carrying out abnormal data correction processing to obtain corrected data;
and performing complement missing data processing on the corrected data based on a difference method.
5. The remote fault diagnosis pushing method according to claim 1, wherein the feature extraction of the signal based on the time domain analysis and the frequency domain analysis comprises:
performing feature extraction on the cleaned state signal by adopting a signal statistical analysis algorithm based on time domain analysis and frequency domain analysis;
wherein, the time domain analysis comprises extracting the minimum value, the maximum value, the average value, the sliding average value, the variance value and the root mean square value of the signal; the frequency domain analysis includes extracting a frequency response peak, a frequency spectrum, a power density spectrum of the signal.
6. The remote fault diagnosis pushing method according to claim 1, wherein the fusing the state feature data of the signals based on the principal analysis algorithm to obtain the running state features of the robot comprises:
respectively carrying out multi-scale decomposition on the state characteristic data of the signal to be fused to obtain a low-pass sub-band component and a band-pass sub-band component of the state characteristic data of the signal to be fused;
fusing the low-pass sub-band components in the state feature data of the signal to be fused based on a first fusion rule of a principal component analysis algorithm to obtain a first fusion feature;
fusing the band-pass sub-band components in the state feature data of the signal to be fused based on a second fusion rule of the principal component analysis algorithm to obtain a second fusion feature;
performing inverse transformation fusion on the first fusion characteristic and the second fusion characteristic to obtain the running state characteristic of the robot;
the first fusion rule is a fusion rule of a principal component analysis algorithm and non-negative matrix factorization; the second fusion rule is a fusion rule of a principal component analysis algorithm and a neural network model.
7. The remote fault diagnosis pushing method according to claim 1, wherein the substituting the running state characteristics of the robot into the fault diagnosis model stored in the fault discrimination model library to determine whether the robot has a fault currently through logic calculation comprises:
and respectively substituting the running state characteristics of the robot into a multi-state fault tree model of the whole robot and a multi-element logistic regression model of the key parts, which are stored in a fault discrimination model library, and judging whether the current whole robot or the key parts of the robot have faults or not through logical calculation.
8. The remote fault diagnosis pushing method according to claim 1, wherein the obtaining of the fault type of the robot and the pushing of the fault type to a management user terminal for real-time display of a fault alarm comprise:
and acquiring the fault type of the robot as a complete machine fault or a key part fault, and pushing the fault type to a corresponding management user terminal to display a fault alarm in real time based on an HTTP (hyper text transport protocol).
9. The remote fault diagnosis pushing method according to claim 1, characterized in that the method further comprises:
matching in a maintenance knowledge base according to the fault types of the robot to obtain similar fault types matched with each other;
performing maintenance information retrieval in the maintenance knowledge base by using the mutually matched similar fault types, and sequencing retrieval results to obtain the sequenced maintenance information of the mutually matched similar fault types;
and pushing the sequenced maintenance information of the similar fault types matched with each other to a corresponding management user terminal based on an HTTP transmission protocol for maintenance suggestion display.
10. A remote troubleshooting pushing device of a robot, the device comprising:
a data acquisition module: the robot state signal acquisition device is used for acquiring a robot state signal and a working environment state signal based on each sensor;
a data preprocessing module: the system comprises a robot state signal acquisition unit, a robot state signal acquisition unit and a working environment state signal acquisition unit, wherein the robot state signal acquisition unit is used for acquiring a robot state signal and a working environment state signal;
a feature extraction module: the system comprises a signal processing module, a signal analyzing module and a signal analyzing module, wherein the signal processing module is used for carrying out characteristic extraction on a signal based on time domain analysis and frequency domain analysis to obtain state characteristic data of the signal;
a feature fusion module: the system is used for fusing state characteristic data of the signals based on a principal analysis algorithm to obtain the running state characteristics of the robot;
and a fault judging module: the fault diagnosis module is used for substituting the running state characteristics of the robot into a fault diagnosis model stored in a fault judgment model library and judging whether the robot has a fault at present through logic calculation;
failure early warning propelling movement module: and the fault type display module is used for acquiring the fault type of the robot and pushing the fault type to a management user terminal for real-time display of fault alarm if the fault is judged to occur.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911114796.2A CN110815224A (en) | 2019-11-14 | 2019-11-14 | Remote fault diagnosis pushing method and device for robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911114796.2A CN110815224A (en) | 2019-11-14 | 2019-11-14 | Remote fault diagnosis pushing method and device for robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110815224A true CN110815224A (en) | 2020-02-21 |
Family
ID=69555200
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911114796.2A Pending CN110815224A (en) | 2019-11-14 | 2019-11-14 | Remote fault diagnosis pushing method and device for robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110815224A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111256814A (en) * | 2020-03-13 | 2020-06-09 | 天津商业大学 | Tower monitoring system and method |
| CN111579001A (en) * | 2020-06-02 | 2020-08-25 | 珠海格力智能装备有限公司 | Fault detection method and device for robot |
| CN111805544A (en) * | 2020-07-10 | 2020-10-23 | 珠海格力智能装备有限公司 | Robot control method and device |
| CN112329825A (en) * | 2020-10-23 | 2021-02-05 | 贵州电网有限责任公司 | Transformer mechanical fault diagnosis method based on information dimension division and decision tree lifting |
| CN112547352A (en) * | 2020-11-06 | 2021-03-26 | 山东农业大学 | Automatic spraying monitoring teleoperation method and system based on digital twinning |
| CN113468210A (en) * | 2021-06-08 | 2021-10-01 | 上海交通大学 | Robot fault diagnosis method and system based on characteristic engineering |
| CN113670645A (en) * | 2021-09-01 | 2021-11-19 | 上海节卡机器人科技有限公司 | Fault detection method and device, electronic equipment and storage medium |
| CN114779615A (en) * | 2022-06-17 | 2022-07-22 | 深圳市捷牛智能装备有限公司 | Robot management and control method and system based on artificial intelligence |
| CN116071902A (en) * | 2023-04-07 | 2023-05-05 | 山东金宇信息科技集团有限公司 | Method, equipment and medium for monitoring power equipment of machine room |
| CN116394263A (en) * | 2023-06-01 | 2023-07-07 | 佛山智能装备技术研究院 | ASO-SVDD-based multi-objective optimization industrial robot anomaly detection method |
| CN116520938A (en) * | 2023-05-09 | 2023-08-01 | 江苏泽润新能科技股份有限公司 | Frequency and amplitude adjustable sine wave generating circuit operation and maintenance system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1472671A (en) * | 2003-06-19 | 2004-02-04 | 上海交通大学 | Method for apparatus status monitoring and performance degenerate fore casting based on network |
| CN103954450A (en) * | 2014-05-19 | 2014-07-30 | 重庆交通大学 | Bearing life degradation performance evaluation index construction method based on main component analysis |
| KR20150059713A (en) * | 2013-11-22 | 2015-06-02 | 한양로보틱스 주식회사 | Remote controllable smart take-out robot central control system through network |
| CN106981057A (en) * | 2017-03-24 | 2017-07-25 | 中国人民解放军国防科学技术大学 | A kind of NSST image interfusion methods based on RPCA |
| US20180087790A1 (en) * | 2016-09-28 | 2018-03-29 | Johnson Controls Technology Company | Systems and methods for automatically creating and using adaptive pca models to control building equipment |
| CN110006469A (en) * | 2019-02-23 | 2019-07-12 | 晋能大土河热电有限公司 | A kind of coal conveyer belt crusing robot and it is operating abnormally recognition methods |
-
2019
- 2019-11-14 CN CN201911114796.2A patent/CN110815224A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1472671A (en) * | 2003-06-19 | 2004-02-04 | 上海交通大学 | Method for apparatus status monitoring and performance degenerate fore casting based on network |
| KR20150059713A (en) * | 2013-11-22 | 2015-06-02 | 한양로보틱스 주식회사 | Remote controllable smart take-out robot central control system through network |
| CN103954450A (en) * | 2014-05-19 | 2014-07-30 | 重庆交通大学 | Bearing life degradation performance evaluation index construction method based on main component analysis |
| US20180087790A1 (en) * | 2016-09-28 | 2018-03-29 | Johnson Controls Technology Company | Systems and methods for automatically creating and using adaptive pca models to control building equipment |
| CN106981057A (en) * | 2017-03-24 | 2017-07-25 | 中国人民解放军国防科学技术大学 | A kind of NSST image interfusion methods based on RPCA |
| CN110006469A (en) * | 2019-02-23 | 2019-07-12 | 晋能大土河热电有限公司 | A kind of coal conveyer belt crusing robot and it is operating abnormally recognition methods |
Non-Patent Citations (1)
| Title |
|---|
| 侯智等: "工业机器人远程监控与故障诊断研究综述", 《机床与液压》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111256814A (en) * | 2020-03-13 | 2020-06-09 | 天津商业大学 | Tower monitoring system and method |
| CN111579001A (en) * | 2020-06-02 | 2020-08-25 | 珠海格力智能装备有限公司 | Fault detection method and device for robot |
| CN111805544A (en) * | 2020-07-10 | 2020-10-23 | 珠海格力智能装备有限公司 | Robot control method and device |
| CN112329825A (en) * | 2020-10-23 | 2021-02-05 | 贵州电网有限责任公司 | Transformer mechanical fault diagnosis method based on information dimension division and decision tree lifting |
| CN112329825B (en) * | 2020-10-23 | 2022-12-06 | 贵州电网有限责任公司 | Transformer mechanical fault diagnosis method based on information dimension division and decision tree lifting |
| CN112547352B (en) * | 2020-11-06 | 2022-08-02 | 山东农业大学 | Automatic spraying monitoring teleoperation method and system based on digital twinning |
| CN112547352A (en) * | 2020-11-06 | 2021-03-26 | 山东农业大学 | Automatic spraying monitoring teleoperation method and system based on digital twinning |
| CN113468210A (en) * | 2021-06-08 | 2021-10-01 | 上海交通大学 | Robot fault diagnosis method and system based on characteristic engineering |
| CN113670645A (en) * | 2021-09-01 | 2021-11-19 | 上海节卡机器人科技有限公司 | Fault detection method and device, electronic equipment and storage medium |
| CN114779615A (en) * | 2022-06-17 | 2022-07-22 | 深圳市捷牛智能装备有限公司 | Robot management and control method and system based on artificial intelligence |
| CN116071902A (en) * | 2023-04-07 | 2023-05-05 | 山东金宇信息科技集团有限公司 | Method, equipment and medium for monitoring power equipment of machine room |
| CN116071902B (en) * | 2023-04-07 | 2023-06-13 | 山东金宇信息科技集团有限公司 | Method, equipment and medium for monitoring power equipment of machine room |
| CN116520938A (en) * | 2023-05-09 | 2023-08-01 | 江苏泽润新能科技股份有限公司 | Frequency and amplitude adjustable sine wave generating circuit operation and maintenance system |
| CN116520938B (en) * | 2023-05-09 | 2023-10-20 | 江苏泽润新能科技股份有限公司 | Frequency and amplitude adjustable sine wave generating circuit operation and maintenance system |
| CN116394263A (en) * | 2023-06-01 | 2023-07-07 | 佛山智能装备技术研究院 | ASO-SVDD-based multi-objective optimization industrial robot anomaly detection method |
| CN116394263B (en) * | 2023-06-01 | 2023-08-25 | 佛山智能装备技术研究院 | ASO-SVDD-based multi-objective optimization industrial robot anomaly detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110815224A (en) | Remote fault diagnosis pushing method and device for robot | |
| CN110861123A (en) | Method and device for visually monitoring and evaluating running state of robot | |
| CN110839016B (en) | Abnormal flow monitoring method, device, equipment and storage medium | |
| CN108444708B (en) | Method for establishing rolling bearing intelligent diagnosis model based on convolutional neural network | |
| CN111368890A (en) | Fault detection method and device and information physical fusion system | |
| CN111947928A (en) | Multi-source information fusion bearing fault prediction system and method | |
| CN111027615A (en) | Middleware fault early warning method and system based on machine learning | |
| CN114265882A (en) | Method, system, device and medium for detecting time sequence signal point abnormity | |
| CN116611013A (en) | Anomaly detection and root cause analysis method and system for industrial time series data | |
| CN115664038B (en) | Intelligent power distribution operation and maintenance monitoring system for electrical safety management | |
| CN116089218A (en) | Dynamic baseline alarm method and system based on Kubernetes historical data and trend analysis | |
| CN114371353A (en) | Power equipment abnormity monitoring method and system based on voiceprint recognition | |
| CN115437358A (en) | Intelligent state monitoring and fault diagnosis system and fault diagnosis method for industrial robot | |
| CN113984389B (en) | Rolling bearing fault diagnosis method based on multiple receptive fields and improved capsule graph neural network | |
| CN117251812A (en) | High-voltage power line operation fault detection method based on big data analysis | |
| CN113092044B (en) | Rotary machine fault diagnosis method based on weighted level visible graph | |
| RU121944U1 (en) | AUTOMATED SYSTEM FOR IDENTIFYING FAULT SENSORS AMONG SENSORS INTENDED FOR CONTROL OF TECHNOLOGICAL PROCESSES | |
| CN113702861B (en) | Power failure position prediction method and device based on big data analysis | |
| CN115169650B (en) | Equipment health prediction method for big data analysis | |
| CN117218405A (en) | Digital fault diagnosis method for battery production equipment | |
| CN108170825B (en) | Distributed energy data monitoring and cleaning method based on cloud platform | |
| CN115081741A (en) | Natural gas metrological verification intelligent prediction method based on neural network | |
| CN109462242B (en) | Power system low-frequency oscillation detection method based on IIR digital filtering and ESPRIT identification algorithm | |
| CN113746862A (en) | Abnormal flow detection method, device and equipment based on machine learning | |
| CN112486096A (en) | Machine tool operation state monitoring method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200221 |
|
| RJ01 | Rejection of invention patent application after publication |