CN110263811B - Equipment running state monitoring method and system based on data fusion - Google Patents
Equipment running state monitoring method and system based on data fusion Download PDFInfo
- Publication number
- CN110263811B CN110263811B CN201910421340.4A CN201910421340A CN110263811B CN 110263811 B CN110263811 B CN 110263811B CN 201910421340 A CN201910421340 A CN 201910421340A CN 110263811 B CN110263811 B CN 110263811B
- Authority
- CN
- China
- Prior art keywords
- equipment
- data
- state
- running state
- distribution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a method and a system for monitoring the running state of equipment based on data fusion, which comprises the following steps: collecting signals of a device sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition; preprocessing signals of a device sensor and defining state data of the device; inputting the well-defined equipment running state data into a GMM model, and outputting equipment running state characteristics and distribution of the state characteristics; analyzing the running state characteristics and distribution of the equipment, and outputting the abnormal state time period of the equipment. The problem that in the prior art, equipment with the characteristics of nonlinearity, isomerism, redundancy and large scale difference for signal features cannot achieve a good monitoring effect can be solved.
Description
Technical Field
The invention relates to the field of equipment monitoring of industrial processes, in particular to a system and a method for monitoring an equipment running state based on data fusion.
Background
With the rapid development of micro-electromechanical system technology, wireless communication technology and sensor technology in recent years, the sensor network has the advantages of rapid construction, convenient deployment and the like, and can finish large-scale and multi-mode information acquisition and automatic monitoring of objects. In addition, mechanical equipment is gradually enlarged and integrated, the working state changes frequently, and other equipment is seriously interfered, so that a vibration signal has strong nonlinearity, non-Gaussian and dynamic properties, the extracted signal characteristics also have the characteristics of nonlinearity, isomerism, redundancy, large scale difference and the like, the monitoring performance of the existing detection and analysis method is not ideal, and the research on the operation monitoring and diagnosis of the nonlinear, non-Gaussian and multi-modal equipment is less.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a method based on data fusion and probabilistic clustering, which obtains operating state data of a device through sensor data, and obtains operating state features and distribution of the state features of the device through a model, so as to find an abnormal state in the operation of the device, so as to solve the problem in the prior art that a device having characteristics of nonlinearity, heterogeneity, redundancy and large scale difference with respect to signal features cannot obtain a good monitoring effect.
The embodiment of the invention provides a method for monitoring the running state of equipment based on data fusion, which is characterized by comprising the following steps:
collecting signals of a device sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition;
preprocessing signals of a device sensor and defining state data of the device;
inputting the well-defined equipment running state data into a GMM model, and outputting equipment running state characteristics and distribution of the state characteristics;
analyzing the running state characteristics and distribution of the equipment, and outputting the abnormal state time period of the equipment.
Specifically, the specific implementation method for preprocessing the signal of the device sensor and defining the state data of the device is as follows:
normalizing the current time values of all the running state sensors of the equipment at any time, and calculating a normalization result; wherein the normalized result is (current time value-historical minimum)/(historical maximum-historical minimum);
and combining the normalized results of all the running state sensors of the equipment at any moment to serve as the state data of the equipment at the current moment.
Preferably, the preprocessing operation further comprises data cleansing.
Specifically, the specific implementation method for inputting the well-defined device operation state data into the GMM model and outputting the device operation state characteristics and the distribution of the state characteristics includes:
1) data set segmentation:
defining the function split _ dataset (A, n),
the input is a data set A, the number of submodels is n, for the data set A, a data set with the length of a preset proportion is randomly selected and output, and the output is a self data set [ A _1, A _2, …, A _ n ];
2) generating a model:
defining the function cluster action (A _ i, stop _ weight),
the input is a data set A _ i, and the stop splitting weight stop _ weight;
inputting the data set A _ i into a standard GMM model to obtain probability distribution M _ i;
outputting predicted values of the pairs A _ i and M _ i, and when the predicted values are 0, classifying the data sets A _ i into A _ i0, and when the predicted values are 1, classifying the data sets A _ i into A _ i 1; clustering A _ i0 to the left node of the probability distribution M _ i when the weight of A _ i0 is greater than stop splitting weight stop _ weight; clustering A _ i1 to the right node of the probability distribution M _ i when the weight of A _ i1 is greater than stop splitting weight stop _ weight;
outputting the final probability distribution M _ i; wherein, the weight calculation mode of the A _ i0 is that the maximum value of the data set A _ i subtracts the minimum value;
3) prediction information:
define function predict _ action (t):
the input is prediction data T, and for each M _ i, the serial number [ result ] _ i of the state feature to which the data belongs and the probability [ score ] _ i of a certain data belonging to the state feature are calculated;
the final output result is result [ [ result ] _1, [ result ] _2, …, [ result ] _ n ], [ [ score ] _1, [ score ] _2, …, [ score ] _ n ].
Specifically, the specific implementation method for analyzing the operating state characteristics and the distribution thereof and outputting the abnormal state time period of the device is as follows:
and the ID of the equipment historical data is taken as a horizontal axis, the state characteristic number of the data is taken as a vertical axis, and the time period when the equipment state is obviously abnormal is quickly positioned.
Preferably, after the abnormal state period of the output device, further comprising:
and displaying abnormal information and sending the information to the communication equipment of the relevant manager.
Preferably, after the abnormal state period of the output device, further comprising:
the period of abnormality information is sent to one or more processors that determine a repair action recommendation for the respective device.
The embodiment of the invention provides a system for monitoring the running state of equipment based on data fusion, which is characterized by comprising the following components:
the signal collection module is used for collecting signals of the equipment sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition;
the preprocessing module is used for preprocessing signals of the equipment sensor and defining state data of the equipment;
the device running state monitoring module is used for inputting the defined device running state data into the GMM model and outputting the device running state characteristics and the distribution of the state characteristics;
and the equipment abnormity output module is used for analyzing the running state characteristics and the distribution of the equipment and outputting the abnormal state time period of the equipment.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic diagram of a method for monitoring an operating state of a device based on data fusion according to a first embodiment of the present invention;
fig. 2 is a schematic diagram of a system for monitoring an operation state of a device based on data fusion according to a second embodiment of the present invention;
fig. 3 is a schematic diagram of a result of monitoring an operation state of the device according to the first embodiment of the present 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.
For the convenience of clearly describing the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", and the like are used to distinguish the same items or similar items with basically the same functions or actions, and those skilled in the art can understand that the words "first", "second", and the like do not limit the quantity and execution order.
Referring to fig. 1, in an embodiment of the present invention, a method for monitoring an operating state of a device based on data fusion is provided,
s101, collecting signals of a device sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition;
in an actual industrial process, the sensor signals are collected according to actual conditions, taking the signals as an example:
the basic information refers to nameplate information and is used for quickly identifying equipment, including device models, device descriptions, manufacturers, port numbers, hardware versions, firmware versions, software versions, IP addresses, MAC addresses and the like;
a health state comprising: power supply, CPU, software, memory, storage, clock, etc.;
the communication state comprises: port, optical power, traffic status, etc.;
the behavioral actions include: recording in real time and monitoring in real time;
the equipment parameters comprise equipment parameters and parameter states;
environmental conditions include harsh conditions as well as normal conditions.
A typical sensor of a device comprises: the system comprises an online infrared thermal imager, visible light video monitoring equipment, a vibration measuring instrument, an ultrasonic thickness gauge, a portable hardness tester, a gas detector, a motor fault diagnosis instrument, a thermal/calibration/electrical measuring instrument and the like.
Multi-level operation is established for various data, and for a multi-source data acquisition layer, data acquisition can be carried out on a plurality of data sources; the data interface layer is connected with the multi-source data acquisition layer, is respectively provided with a plurality of data interfaces corresponding to various data types, and receives multi-source data by using the data interfaces of the corresponding data types; and the big data fusion management layer is used for respectively storing the data received by the data interface layer, calculating and analyzing the big data and the like.
Preferably, the data interface layer comprises a SQOOP interface for transmitting relational data, a JMS interface for transmitting real-time data stream, and an FTP interface for transmitting regular data;
the big data fusion management layer comprises a data warehouse tool Hive, an Hbase database, an SQL big data query tool Impala and a Hadoop resource manager YARN;
the system also comprises a large-scale Streaming data processing framework Spark Streaming, a language pack Spark R for using Apache Spark from R language, a machine learning algorithm library MLlib, Spark GraphX and Python programming environments for executing graph parallel computing.
Preferably, the system also comprises an open source algorithm library, such as a space-time geographic analysis module, an influence range analysis module, a graph calculation analysis module and an association relation analysis module;
preferably, the model access interface layer comprises: the Java database connection interface JDBC is used for executing SQL sentences; the open database is connected with ODBC and used for supporting SQL language; spark SQL, used to realize the calculation of the structured data; and the HUE interface is used for realizing development and debugging of a Hadoop ecosystem.
The multi-level operation enables the equipment management personnel to flexibly and accurately process various types of signal data, and the existing statistical analysis tools are conveniently and rapidly adopted to carry out rapid modeling, so that simple or complex analysis results are generated.
S102, preprocessing signals of a device sensor and defining state data of the device;
the specific implementation method comprises the following steps:
normalizing the current time values of all the running state sensors of the equipment at any time, and calculating a normalization result; wherein the normalized result is (current time value-historical minimum)/(historical maximum-historical minimum);
and combining the normalized results of all the running state sensors of the equipment at any moment to serve as the state data of the equipment at the current moment.
The normalized result is stored as an array, matrix information and the like according to a preset rule, and each model is convenient to adopt and use.
Preferably, a data cleansing process is performed prior to normalization, as is practical.
S103, inputting the well-defined equipment running state data into the GMM model, and outputting equipment running state characteristics and distribution of the state characteristics;
the specific implementation method comprises the following steps:
1) data set segmentation:
defining the function split _ dataset (A, n),
the input is a data set A, the number of submodels is n, for the data set A, a data set with the length of a preset proportion is randomly selected and output, and the output is a self data set [ A _1, A _2, …, A _ n ];
2) generating a model:
defining the function cluster action (A _ i, stop _ weight),
the input is a data set A _ i, and the stop splitting weight stop _ weight;
inputting the data set A _ i into a standard GMM model to obtain probability distribution M _ i;
outputting predicted values of the pairs A _ i and M _ i, and when the predicted values are 0, classifying the data sets A _ i into A _ i0, and when the predicted values are 1, classifying the data sets A _ i into A _ i 1; clustering A _ i0 to the left node of the probability distribution M _ i when the weight of A _ i0 is greater than stop splitting weight stop _ weight; clustering A _ i1 to the right node of the probability distribution M _ i when the weight of A _ i1 is greater than stop splitting weight stop _ weight;
outputting the final probability distribution M _ i; wherein, the weight calculation mode of the A _ i0 is that the maximum value of the data set A _ i subtracts the minimum value;
3) prediction information:
define function predict _ action (t):
the input is prediction data T, and for each M _ i, the serial number [ result ] _ i of the state feature to which the data belongs and the probability [ score ] _ i of a certain data belonging to the state feature are calculated;
the final output result is result [ [ result ] _1, [ result ] _2, …, [ result ] _ n ], [ [ score ] _1, [ score ] _2, …, [ score ] _ n ].
The standard GMM Model, which is a popular algorithm, is a Gaussian Mixture Model (Gaussian Mixture Model), which assumes that data obeys the Mixture Gaussian Distribution, in other words, the data can be regarded as being generated from several Gaussian distributions. In fact, from the central limit theorem, it can be seen that the assumption of Gaussian distribution is rather reasonable, and besides, Gaussian distribution has some very good properties in computation. In addition, the texture Model itself can become arbitrarily complex, and by increasing the number of models, we can arbitrarily approximate any continuous probability density distribution.
Each GMM consists of a distribution of gaussians, each known as a "Component", which when linearly added together make up the probability density function of the GMM:
from the above equation, if we want to randomly take a point from the distribution of the GMM, we can actually divide into two steps: first randomly selecting one among the components, the probability of each Component being selected is actually its coefficient, and after selecting a Component, then considering selecting a point from the distribution of the Component alone-here having returned to the ordinary Gaussian distribution, translated to a known problem.
The GMM is adopted for clustering, and for the existing data, the probability distribution of the GMM can be obtained by only deducing the probability distribution of the GMM according to the data, and then the Component of the GMM actually corresponds to the cluster. Estimating probability densities from data is commonly referred to as density estimation, in particular when we know (or assume) the form of the probability density function, and the process in which the parameters are to be estimated is referred to as "parameter estimation".
Now assuming that there are data points and that they obey a certain distribution (denoted), the values of some of the parameters inside are now to be determined, e.g. in GMM, and need to be determined. Finding the set of parameters whose determined probability distribution yields the maximum probability for the given data points, which probability is substantially equal, calls this product as a Likelihood Function (likehood Function). The probability of a single point is usually small, and many small numbers are multiplied together to easily cause floating point underflow in a computer, so that the logarithm of the floating point underflow is usually taken, and the products are summed to obtain the log-likelihood function. The parameter estimation process is then completed by maximizing the function (usually by deriving and making the derivative equal to zero and then solving the equation), i.e. finding a set of parameter values that maximizes the likelihood function, which is then considered to be the most appropriate parameter.
The log-likelihood function of the GMM is seen below:
because the addition exists in the logarithmic function, the maximum value cannot be directly obtained by directly solving the equation by means of derivation. To solve this problem, a random selection of points from the GMM was previously taken: the method is divided into two steps,
estimate the probability that data is generated by each Component (not the probability that each Component is selected): for each datum, its probability generated by the first Component is
Since the sum in the equation is also the value to be estimated, using an iterative method, the sum is assumed to be known at the time of calculation, and the value (or initial value) obtained from the last iteration will be taken.
Estimating parameters of each Component: now, it is assumed that the "probability that data is generated by Component" obtained in the previous step is also used as a contribution of the Component in generating the data, or the value can be regarded as the value in which the part is generated by Component. Considering all data points collectively, these points can now be considered as being generated by Component in practice. Since each Component is a standard Gaussian distribution, the parameter values corresponding to the maximum likelihood can be easily distributed:
among these, and also in a rational manner, can be estimated.
The first two steps are iterated repeatedly until the values of the likelihood functions converge.
Therefore, the observation probability of each operation state characteristic of the equipment can be obtained by improving the GMM model, namely the probability distribution of the operation state characteristic of the equipment is output.
And S104, analyzing the running state characteristics and distribution of the equipment, and outputting the abnormal state time period of the equipment.
As can be seen from fig. 3, one implementation method is to quickly locate a time period in which the device status is obviously abnormal, with the ID of the device history data as the horizontal axis and the status feature number of the data as the vertical axis, where the time period of the abnormality in fig. 3 is shown as a black box.
For those skilled in the art, any method for comparing the current device status feature distribution with the historical device status feature is suitable for analyzing the abnormal status period result.
Preferably, after the abnormal state period of the output device, further comprising:
and displaying abnormal information and sending the information to the communication equipment of the relevant manager.
Preferably, after the abnormal state period of the output device, further comprising:
the period of abnormality information is sent to one or more processors that determine a repair action recommendation for the respective device.
The embodiment of the invention provides a data fusion-based equipment running state monitoring system, which comprises:
the signal collection module is used for collecting signals of the equipment sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition;
the preprocessing module is used for preprocessing signals of the equipment sensor and defining state data of the equipment;
the device running state monitoring module is used for inputting the defined device running state data into the GMM model and outputting the device running state characteristics and the distribution of the state characteristics;
and the equipment abnormity output module is used for analyzing the running state characteristics and the distribution of the equipment and outputting the abnormal state time period of the equipment.
In the technical scheme of the invention, the modules are all common hardware processing devices, such as an MCU, a CPU, an FPGA, an image processor GPU and the like.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (6)
1. A method for monitoring the running state of equipment based on data fusion is characterized by comprising the following steps:
collecting signals of a device sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition;
preprocessing signals of a device sensor and defining state data of the device;
inputting the well-defined equipment running state data into a GMM model, and outputting equipment running state characteristics and distribution of the state characteristics;
analyzing the running state characteristics and the distribution of the equipment, and outputting the abnormal state time period of the equipment;
the specific implementation method for inputting the defined equipment running state data into the GMM model and outputting the equipment running state characteristics and the distribution of the state characteristics comprises the following steps:
1) data set segmentation:
defining the function split _ dataset (A, n),
the input is a data set A, the number of submodels is n, for the data set A, a data set with the length of a preset proportion is randomly selected and output, and the data set A is output as a set [ A _1, A _2, …, A _ n ] of the data set;
2) generating a model:
defining the function cluster action (A _ i, stop _ weight),
the input is a data set A _ i, and the stop splitting weight stop _ weight;
inputting the data set A _ i into a standard GMM model to obtain probability distribution M _ i;
outputting predicted values of the pairs A _ i and M _ i, and when the predicted values are 0, classifying the data sets A _ i into A _ i0, and when the predicted values are 1, classifying the data sets A _ i into A _ i 1; clustering A _ i0 to the left node of the probability distribution M _ i when the weight of A _ i0 is greater than the stop splitting weight stop _ weight; clustering A _ i1 to the right node of the probability distribution M _ i when the weight of A _ i1 is greater than stop splitting weight stop _ weight;
outputting the final probability distribution M _ i; wherein, the weight calculation mode of the A _ i0 is that the maximum value of the data set A _ i subtracts the minimum value;
3) prediction information:
define function predict _ action (t):
the input is prediction data T, and for each M _ i, the serial number [ result ] _ i of the state feature to which the data belongs and the probability [ score ] _ i of a certain data belonging to the state feature are calculated;
the final output result is result [ [ result ] _1, [ result ] _2, …, [ result ] _ n ], [ [ score ] _1, [ score ] _2, …, [ score ] _ n ].
2. The apparatus operating condition monitoring method according to claim 1,
the specific implementation method for preprocessing the signals of the equipment sensor and defining the state data of the equipment comprises the following steps:
normalizing the current time values of all the running state sensors of the equipment at any time, and calculating a normalization result; wherein the normalized result is (current time value-historical minimum)/(historical maximum-historical minimum);
and combining the normalized results of all the running state sensors of the equipment at any moment to serve as the state data of the equipment at the current moment.
3. The apparatus operating condition monitoring method according to claim 1,
the specific implementation method for analyzing the running state characteristics and the distribution of the running state characteristics and outputting the abnormal state time interval of the equipment comprises the following steps:
and the ID of the equipment historical data is taken as a horizontal axis, the state characteristic number of the data is taken as a vertical axis, and the time period when the equipment state is obviously abnormal is quickly positioned.
4. The apparatus operating condition monitoring method according to claim 1, further comprising, after the abnormal condition period of the output apparatus:
and displaying abnormal information and sending the information to the communication equipment of the relevant manager.
5. The apparatus operating condition monitoring method according to claim 1, further comprising, after the abnormal condition period of the output apparatus:
the period of abnormality information is sent to one or more processors that determine a repair action recommendation for the respective device.
6. An equipment operation state monitoring system based on data fusion, which is characterized in that the equipment operation state monitoring method of any one of claims 1-5 is adopted, and comprises the following steps:
the signal collection module is used for collecting signals of the equipment sensor; wherein the signal comprises basic information of the equipment, health state, communication state, behavior action, equipment parameter and environmental condition;
the preprocessing module is used for preprocessing signals of the equipment sensor and defining state data of the equipment;
the device running state monitoring module is used for inputting the defined device running state data into the GMM model and outputting the device running state characteristics and the distribution of the state characteristics;
and the equipment abnormity output module is used for analyzing the running state characteristics and the distribution of the equipment and outputting the abnormal state time period of the equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910421340.4A CN110263811B (en) | 2019-05-21 | 2019-05-21 | Equipment running state monitoring method and system based on data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910421340.4A CN110263811B (en) | 2019-05-21 | 2019-05-21 | Equipment running state monitoring method and system based on data fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110263811A CN110263811A (en) | 2019-09-20 |
| CN110263811B true CN110263811B (en) | 2021-03-26 |
Family
ID=67914916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910421340.4A Active CN110263811B (en) | 2019-05-21 | 2019-05-21 | Equipment running state monitoring method and system based on data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110263811B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111077876B (en) * | 2019-12-11 | 2021-06-08 | 湖南大唐先一科技有限公司 | Power station equipment state intelligent evaluation and early warning method, device and system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009175929A (en) * | 2008-01-23 | 2009-08-06 | Toyota Central R&D Labs Inc | Driver condition estimating device and program |
| CN102361014A (en) * | 2011-10-20 | 2012-02-22 | 上海大学 | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process |
| WO2014092946A1 (en) * | 2012-12-14 | 2014-06-19 | Alcatel-Lucent Usa Inc. | Method and system for disaggregating thermostatically controlled appliance energy usage from other energy usage |
| CN104595170A (en) * | 2014-12-18 | 2015-05-06 | 中国矿业大学 | Air compressor monitoring diagnosis system and method adopting adaptive kernel Gaussian hybrid model |
| CN104902220A (en) * | 2014-12-09 | 2015-09-09 | 哈尔滨理工大学 | Equipment running state cloud monitoring device based on improved GMM (Gaussian Mixture Model) method |
| CN106651906A (en) * | 2015-10-30 | 2017-05-10 | 国网山西省电力公司电力科学研究院 | Test method for motion characteristic of high-voltage breaker based on improved Gaussian mixture model |
| CN106650122A (en) * | 2016-12-27 | 2017-05-10 | 宝鸡文理学院 | Equipment variable working condition operation risk evaluation method |
| CN107478273A (en) * | 2017-08-14 | 2017-12-15 | 武汉科技大学 | Based on embedded and technology of Internet of things sewer monitoring system and method |
| CN108199795A (en) * | 2017-12-29 | 2018-06-22 | 北京百分点信息科技有限公司 | The monitoring method and device of a kind of equipment state |
| CN109029992A (en) * | 2018-06-15 | 2018-12-18 | 深圳大学 | The method and system of the bearing failure diagnosis of machine tool chief axis |
| CN109405889A (en) * | 2018-12-13 | 2019-03-01 | 中国航空工业集团公司上海航空测控技术研究所 | System and method for for the prediction of development machine working arm decelerator trouble |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10430943B2 (en) * | 2016-10-07 | 2019-10-01 | Sony Corporation | Automated nuclei area/number estimation for IHC image analysis |
| CN107610464B (en) * | 2017-08-11 | 2019-02-12 | 河海大学 | A kind of trajectory predictions method based on Gaussian Mixture time series models |
| CN109702744A (en) * | 2019-01-15 | 2019-05-03 | 北京工业大学 | A method of the robot learning by imitation based on dynamic system model |
-
2019
- 2019-05-21 CN CN201910421340.4A patent/CN110263811B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009175929A (en) * | 2008-01-23 | 2009-08-06 | Toyota Central R&D Labs Inc | Driver condition estimating device and program |
| CN102361014A (en) * | 2011-10-20 | 2012-02-22 | 上海大学 | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process |
| WO2014092946A1 (en) * | 2012-12-14 | 2014-06-19 | Alcatel-Lucent Usa Inc. | Method and system for disaggregating thermostatically controlled appliance energy usage from other energy usage |
| CN104902220A (en) * | 2014-12-09 | 2015-09-09 | 哈尔滨理工大学 | Equipment running state cloud monitoring device based on improved GMM (Gaussian Mixture Model) method |
| CN104595170A (en) * | 2014-12-18 | 2015-05-06 | 中国矿业大学 | Air compressor monitoring diagnosis system and method adopting adaptive kernel Gaussian hybrid model |
| CN106651906A (en) * | 2015-10-30 | 2017-05-10 | 国网山西省电力公司电力科学研究院 | Test method for motion characteristic of high-voltage breaker based on improved Gaussian mixture model |
| CN106650122A (en) * | 2016-12-27 | 2017-05-10 | 宝鸡文理学院 | Equipment variable working condition operation risk evaluation method |
| CN107478273A (en) * | 2017-08-14 | 2017-12-15 | 武汉科技大学 | Based on embedded and technology of Internet of things sewer monitoring system and method |
| CN108199795A (en) * | 2017-12-29 | 2018-06-22 | 北京百分点信息科技有限公司 | The monitoring method and device of a kind of equipment state |
| CN109029992A (en) * | 2018-06-15 | 2018-12-18 | 深圳大学 | The method and system of the bearing failure diagnosis of machine tool chief axis |
| CN109405889A (en) * | 2018-12-13 | 2019-03-01 | 中国航空工业集团公司上海航空测控技术研究所 | System and method for for the prediction of development machine working arm decelerator trouble |
Non-Patent Citations (2)
| Title |
|---|
| Classification of operating conditions of wind turbines for a class-wise condition monitoring strategy;Jong M.Ha 等;《Renewable Energy》;20161102;第103卷;全文 * |
| 基于KPCA-GMM的球磨机状态监测与评估研究;王恒 等;《矿山机械》;20140910;第42卷(第9期);第66-70页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110263811A (en) | 2019-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107086944B (en) | Anomaly detection method and device | |
| CN110851338B (en) | Abnormality detection method, electronic device, and storage medium | |
| CN111124840B (en) | Method and device for predicting alarm in business operation and maintenance and electronic equipment | |
| US11048729B2 (en) | Cluster evaluation in unsupervised learning of continuous data | |
| CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
| CN113518011B (en) | Abnormality detection method and apparatus, electronic device, and computer-readable storage medium | |
| CN113792453A (en) | Partial discharge monitoring system, method and device based on digital twins | |
| CN111045894A (en) | Database anomaly detection method and device, computer equipment and storage medium | |
| CN107766573B (en) | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium based on data processing | |
| CN117194919A (en) | Production data analysis system | |
| CN108009063B (en) | Method for detecting fault threshold of electronic equipment | |
| US20180307218A1 (en) | System and method for allocating machine behavioral models | |
| JP2003044123A (en) | Plant diagnostic device | |
| CN110263811B (en) | Equipment running state monitoring method and system based on data fusion | |
| WO2018036402A1 (en) | Method and device for determining key variable in model | |
| CN117035563B (en) | Product quality safety risk monitoring method, device, monitoring system and medium | |
| CN112380073B (en) | Fault position detection method and device and readable storage medium | |
| CN117421994A (en) | Edge application health monitoring method and system | |
| CN114139589A (en) | Fault diagnosis method, device, equipment and computer readable storage medium | |
| Barlowe et al. | Multivariate visual explanation for high dimensional datasets | |
| CN111221704B (en) | Method and system for determining running state of office management application system | |
| US11762562B2 (en) | Performance analysis apparatus and performance analysis method | |
| CN113688120A (en) | Quality detection method and device for data warehouse and electronic equipment | |
| CN114238368A (en) | Transformer fault diagnosis method and device, computer equipment and readable storage medium | |
| CN111160715A (en) | BP neural network based new and old kinetic energy conversion performance evaluation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Liu Jian Inventor before: Liu Jian |