CN110298409A - Multi-source data fusion method towards electric power wearable device - Google Patents
Multi-source data fusion method towards electric power wearable device Download PDFInfo
- Publication number
- CN110298409A CN110298409A CN201910596096.5A CN201910596096A CN110298409A CN 110298409 A CN110298409 A CN 110298409A CN 201910596096 A CN201910596096 A CN 201910596096A CN 110298409 A CN110298409 A CN 110298409A
- Authority
- CN
- China
- Prior art keywords
- data
- information
- electric power
- fusion
- wearable device
- 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
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 18
- 230000004927 fusion Effects 0.000 claims abstract description 71
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000011156 evaluation Methods 0.000 claims abstract description 8
- 230000000295 complement effect Effects 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 21
- 238000005516 engineering process Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000013477 bayesian statistics method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001373 regressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The present invention relates to Data fusion technique fields, a kind of multi-source data fusion method towards electric power wearable device is specifically disclosed, include: S10: classification is extracted to the collected data information of the multisensor syste of electric power wearable device, rejects unrelated data information and acquisition needs to carry out the data information of data fusion；S20: by building bayes predictive model to needing the data information for carrying out data fusion to be analyzed and processed；S30: it is analyzed and processed by the Weighted Fusion algorithm based on variance evaluation to sorted data information is extracted.The present invention provides a kind of multi-source data fusion method towards electric power wearable device, has the advantages that efficiently and accurate.
Description
Technical field
The present invention relates to Data fusion technique field more particularly to a kind of multi-source datas towards electric power wearable device
Fusion method.
Background technique
Generally, due to which work about electric power personnel will carry out operation in the power domain of various danger, so being frequently necessary to wear
Wear the electric power wearable device such as Intelligent safety helmet and Intelligent glove.Conventional electric power wearable device all has many biographies
Sensor, in order to realize accurate sensing, it is necessary to which convergence analysis is carried out to the collected data of each sensor.
With the rapid development of big data information technology, information fusion technology is widely applied in field of power communication, more
Data Fusion of Sensor MSDF (Multi-sensor Data Fusion) technology reaches its maturity, for electric power wearable device
Each control system for, sensor states monitoring and the reliability etc. of fault detection, diagnosis and compensation method all have ten
Divide important theoretical significance and application value.Existing research at present proposes much information Fusion Model algorithm and method,
Main algorithm is to know method for distinguishing based on the identification of model class, Parameter analysis and knowledge model class.
Data anastomosing algorithm is the basic content of Data Fusion.It is by multidimensional input data according to data fusion
Function, on different data fusion PCR use different mathematical methods, to data carry out integrated treatment, finally realize number
According to fusion.Has a large amount of data anastomosing algorithm at present, they there are respective advantage and disadvantage, wherein popular data fusion is calculated
Method has bayes method, evidence theory reasoning, fuzzy theory and neural network method etc..According to the reality of electric power field work
Situation and the specific requirement to wearable device acquisition parameter, display Bayes estimation adopt wearable device multisensor
The multi-source information of collection can be merged effectively, and can effectively improve accident analysis data accuracy and feedback information can
By property.
Bayes method is a kind of inference method merged earliest applied to uncertain data, and basic thought is to set
Under conditions of determining prior probability, posterior probability is calculated using Bayes rule, thus made a policy according to posterior probability, this
Sample can handle uncertain problem.The Major Difficulties of bayes method are to find suitable probability distribution, especially work as
Data just seem more difficult when coming from low-grade sensor.
Currently with electric power wearable device carry out electric power field work data acquisition when, the diversity of data it is more and
Staff is higher to the accuracy of data demand, and existing data anastomosing algorithm is not only complicated but also universal, and there is no specially
Door is directed to the data anastomosing algorithm of electric power wearable device, this will lead to, and existing data anastomosing algorithm is wearable to electric power to be set
Not efficient and accurate enough the problem of standby data processed result.
Therefore, it is necessary to develop a kind of multisource data fusion and classification method towards electric power wearable device, in favor of solution
Not efficient and inaccurate processing result enough the problem of certainly existing processing method.
Summary of the invention
It is an advantage of the invention to provide a kind of multi-source data fusion method towards electric power wearable device, tools
Have the advantages that efficient and accurate.
To achieve these objectives, the present invention provides a kind of multi-source data fusion method towards electric power wearable device, packet
It includes:
S10: extracting classification to the collected data information of the multisensor syste of electric power wearable device, rejects
Unrelated data information and acquisition needs to carry out the data information of data fusion；
S20: by building bayes predictive model to needing the data information for carrying out data fusion to be analyzed and processed；
S30: it is carried out at analysis by the Weighted Fusion algorithm based on variance evaluation to sorted data information is extracted
Reason.
Preferably, the S10 includes:
S101: the collected data information of the multisensor syste is divided into redundancy, complementary information and collaboration and is believed
Breath；
S102: the redundancy and complementary information are rejected；
S103: step S20 is executed to the cooperative information.
Preferably, the S20 includes:
S201: optimal bayesian estimation value is obtained comprising:
S2011: establishing n bayes predictive models to be selected to the collected data information of multisensor syste, using each
A bayes predictive model to be selected speculates the probability of sensor failure；
S2012: the n bayes predictive models to be selected add the prediction conclusion of sensor failure
Weight average, resulting weight are the optimal bayesian estimation value；
S202: the optimal bayesian estimation value is compared with preset value, when optimal bayesian estimation value is greater than in advance
If rejecting the data when value.
Preferably, the S30 includes:
S301: the average value of moment multiple sensor signals is calculated；
S302: setting data window length T estimates the variance of each sensor mean deviation signal；
S303: the minimum variance of each sensor is estimated according to the variance of the mean deviation signal；
S304: melted according to the weighting that the minimum variance of each sensor calculates the collected data information of multisensor syste
Conjunction value.
On the other hand, the present invention provides a kind of multisource data fusion device towards electric power wearable device, comprising:
Classification extraction module is carried out for the collected data information of multisensor syste to electric power wearable device
Classification is extracted, unrelated data information is rejected and acquisition needs to carry out the data information of data fusion；
Bayes predictive model constructs module, for by building bayes predictive model to needing to carry out data fusion
Data information be analyzed and processed；
Weighted Fusion module, for being believed by the Weighted Fusion algorithm based on variance evaluation sorted data are extracted
Breath is analyzed and processed.
Preferably, the classification extraction module includes:
Taxon, for the collected data information of the multisensor syste to be divided into redundancy, complementary information
And cooperative information；
Culling unit, for rejecting the redundancy and complementary information；
Selection unit, for being analyzed and processed by constructing bayes predictive model to cooperative information.
Preferably, the bayes predictive model building module includes:
Acquiring unit, for obtaining optimal bayesian estimation value；
Comparing unit, for comparing the optimal bayesian estimation value with preset value, when optimal bayesian estimation
When value is greater than preset value, the data are rejected.
Preferably, the acquiring unit includes:
Presumption units, for establishing n Bayesian forecasting moulds to be selected to the collected data information of multisensor syste
Type speculates the probability of sensor failure using each bayes predictive model to be selected；
It is weighted and averaged unit, for by predictions of the n bayes predictive model to be selected for sensor failure
Conclusion is weighted and averaged, and resulting weight is the optimal bayesian estimation value；
Preferably, the Weighted Fusion module includes:
Average calculation unit, for calculating the average value of moment multiple sensor signals；
Average variance computing unit estimates the side of each sensor mean deviation signal for setting data window length T
Difference；
Minimum variance computing unit, for estimating the minimum of each sensor according to the variance of the mean deviation signal
Variance；
Integrated unit, for calculating the collected data information of multisensor syste according to the minimum variance of each sensor
Weighted Fusion value.
The beneficial effects of the present invention are: for the bayes method, evidence theory reasoning, fuzzy at present than Chang Liuhang
The data anastomosing algorithms such as theoretical and neural network method are studied, data when analyzing them applied to electric power wearable device
The advantage and disadvantage of fusion finally propose the multisource data fusion and classification method that are directed to electric power wearable device.The present invention improves
The accuracy and high efficiency of electric power field work data processing, by the data fusion for designing suitable electric power wearable device
Method improves fusion efficiencies, simplifies unnecessary fusion process.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
Some embodiments of the present invention, for those of ordinary skill in the art, without any creative labor,
It can also be obtained according to these attached drawings other attached drawings.
Fig. 1 is the flow diagram for the multi-source data fusion method towards electric power wearable device that embodiment provides；
Fig. 2 is the flow chart for this prediction model that embodiment provides；
Fig. 3 is the structural block diagram for the multisource data fusion device towards electric power wearable device that embodiment provides.
Specific embodiment
To enable the purpose of the present invention, feature, advantage more obvious and understandable, below in conjunction with of the invention real
The attached drawing in example is applied, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that disclosed below
Embodiment is only a part of the embodiment of the present invention, and not all embodiment.Based on the embodiments of the present invention, this field
Those of ordinary skill's all other embodiment obtained without making creative work, belongs to guarantor of the present invention
The range of shield.
In the description of the present invention, it is to be understood that, when a component is considered as " connection " another component, it
It can be directly to another component or may be simultaneously present the component being centrally located.When a component is considered as
" setting exist " another component, it can be to be set up directly on another component or may be simultaneously present and be centrally located
Component.
In addition, the indicating positions such as term " length " " short " "inner" "outside" or positional relationship are the orientation that is shown based on attached drawing
Or positional relationship, it is merely for convenience of the description present invention, rather than the device or original part of indication or suggestion meaning must have
This specific orientation is operated with specific orientation construction, should not be understood as limitation of the invention with this.
To further illustrate the technical scheme of the present invention below with reference to the accompanying drawings and specific embodiments.
Embodiment one
The present embodiment provides a kind of multi-source data fusion methods towards electric power wearable device, are suitable for power equipment
Electric power field work data-handling efficiency can be improved in application scenarios in field, and the multi-source data fusion method is by one kind
Multisource data fusion device executes, and passes through software and or hardware realization.
Fig. 1 is the flow chart for the multi-source data fusion method that the present embodiment one provides.
The present embodiment provides a kind of multi-source data fusion methods towards electric power wearable device, comprising:
S10: extracting classification to the collected data information of the multisensor syste of electric power wearable device, rejects
Unrelated data information and acquisition needs to carry out the data information of data fusion.
Preferably, the collected information of multisensor syste includes pressure, image, voltage, electric current, temperature, noise etc..
In the present embodiment, S101 includes:
S101: the collected data information of the multisensor syste is divided into redundancy, complementary information and collaboration and is believed
Breath；
S102: the redundancy and complementary information are rejected；
S103: step S20 is executed to the cooperative information.
Preferably, consistency verification of data is carried out to redundancy, unified model is established to complementary information, collaboration is believed
Breath carries out data fusion.
Specifically, redundancy is multiple information of the same feature provided by multiple standalone sensors, need to be to redundancy
Same feature carries out data consistency checks；For example, multiple positions of foot's the same area are provided with foot pressure sensing
Device, the foot pressure information that multiple foot pressure sensors transmit just belong to redundancy, need to carry out consistency check, bright
The numerical value of explict occurrence deviation should be removed.Complementary information is that each sensor provides feature independent of each other, need to be believed complementation
Breath carries out the foundation of unified model；For example, what hand pressure sensor and foot pressure sensor transmitted is pressure independent of each other
Force signal needs to establish unified model to express electric power personnel hand sensor and Foot sensor when a certain posture is presented
The corresponding relationship being likely to occur；Cooperative information is that the acquisition of A sensor information must rely on the information of B sensor, collaboration letter
Breath needs to carry out Data Fusion.
S20: by building bayes predictive model to needing the data information for carrying out data fusion to be analyzed and processed.
In the present embodiment, the S20 includes:
S201: optimal bayesian estimation value is obtained comprising:
S2011: establishing n bayes predictive models to be selected to the collected data information of multisensor syste, using each
A bayes predictive model to be selected speculates the probability of sensor failure；
Specifically, the collected data information of multisensor syste pushes away the n bayes predictive models to be selected built
The unknown phenomenon probability of happening surveyed is unknown.Preferably, the unknown phenomenon can be sensor failure.
S2012: the n bayes predictive models to be selected add the prediction conclusion of sensor failure
Weight average, resulting weight are the optimal bayesian estimation value.
It is understood that bayes predictive model is a kind of prediction carried out with Bayesian statistics.Bayesian statistics
Different from general statistical method, not merely with model information and data information, and prior information is made full use of.Pass through
Bayes predictive model is compared by the method for proof analysis with the prediction result of common regressive prediction model, the results showed that
Bayes predictive model has apparent superiority.In order to facilitate understanding, bayes predictive model is explained below
It is bright:
The present embodiment is defined on the item to be sorted of n node in Bayesian network according to chain type criterion (Chain Rule)
Are as follows: A={ x1, x2 ..., xn }, wherein x is the characteristic attribute of A.
When category set B={ y1, y2 ..., yn } indicates its probability density,
P (x) is prior probability " (Prior probability), i.e., before the generation of B event, to the one of the A probability of happening
A judgement.
P (A | B) be known as " posterior probability " (Posterior probability), i.e., after the generation of B event, to A thing
Part probability reappraises.
P (B | A)/P (B) is known as " plausibility function " (Likelyhood), can adjust the factor and estimate probability, more connect
Nearly true probability.
The item set to be sorted classified known to one is found, statistics obtains the condition of each characteristic attribute under of all categories
Probability.Each characteristic attribute is conditional sampling, then can make following derivation according to Bayes' theorem:
Denominator is constant for all categories in above formula, and molecule maximizes, and obtains information fusion value, output system state
As shown in Figure 2.Basic Bayesian Network frame includes that the expression of network, the modeling of network, network reasoning (are answered
Inquiry problem under any given evidence), many parts such as the prediction of network and diagnosis.
S202: the optimal bayesian estimation value is compared with preset value, when optimal bayesian estimation value is greater than in advance
If rejecting the data when value.
It is understood that the optimal bayesian estimation value is compared with preset value, when optimal bayesian estimation
When value is greater than preset value, it is believed that the corresponding sensor failure of the data should reject the data.
S30: it is carried out at analysis by the Weighted Fusion algorithm based on variance evaluation to sorted data information is extracted
Reason.
In the present embodiment, the S30 includes:
S301: the average value of t moment multiple sensor signals is calculated
Wherein, yitFor the measured value of t moment sensor i.
S302: setting data window length T estimates each sensor mean deviation signal e 'itVariance
S303: each sensor variance is estimated according to the variance of the mean deviation signal
It should be noted that the factor of analyzing influence variance evaluation precision, above formula show varianceEvaluated error byIt determines.Deriving to the distribution of the latter can obtain, mean deviation e 'itTo measure noise eitLine
Property combination, obeyNormal distribution, thenIt obeysGamma distribution.Further byIt can obtain:And meet:
Formula illustrate estimate of varianceFor actual valueUnbiased esti-mator,The formula side of explanation
Poor evaluated error is directly proportional to the size of variance actual value, is inversely proportional with shifting data window length N.
S304: melted according to the weighting that the minimum variance of each sensor calculates the collected data information of multisensor syste
Conjunction value；
Generally, each quantity of state has decoupled in component level fusion, measurement equation expression formula are as follows:
H=Mx+k；
In formula, x is quantity of state to be measured, H=[h1 h2…hn]TMeasurement vector, k=[k are tieed up for n1 k2…kn]TIt ties up and surveys for n
Measure noise vector, kiMutually indepedent and satisfaction For the variance of each signal source；M is known
N × 1 tie up calculation matrix, M=[1...1]T。
Under the criterion of linear minimum-variance estimation, the Weighted Fusion value of the collected data information of multisensor syste
Are as follows:
Wherein, weighting coefficient calculation formula are as follows:
It should be noted that Weighted Fusion error is mean square error, concrete form is as follows:
According to the important conclusion of the available Weighted Fusion of above formula: precision again poor sensor participate in Weighted Fusion can
Improve the precision of fusion results.It should be noted that the premise that above-mentioned conclusion is set up is sensor varianceIt is accurately known, and
Measurement variance is influenced by various factors in practice, estimate of varianceWith actual valueThere are certain evaluated error,
When evaluated error is larger, the correctness of above-mentioned conclusion is influenced whether.Therefore Weighted Fusion precision is practical depends on sensor survey
Measure the accuracy of Noise Variance Estimation.
The present embodiment is for the bayes method than Chang Liuhang, evidence theory reasoning, fuzzy theory and neural network at present
The data anastomosing algorithms such as method are studied, the advantage and disadvantage of data fusion when analyzing them applied to electric power wearable device, most
The multisource data fusion and classification method that are directed to electric power wearable device are proposed afterwards.Embodiment improves electric power field works
The accuracy and high efficiency of data processing, by designing the data fusion method of suitable electric power wearable device, simplification need not
The fusion process wanted, effectively increases fusion efficiencies.
Embodiment two
Multisource data fusion device provided in this embodiment can be used for executing the multi-source data of the embodiment of the present invention offer
Fusion method has corresponding function and beneficial effect.
Fig. 3 is the structural block diagram for the multisource data fusion device that the present embodiment two provides.
Referring to Fig. 3, a kind of multisource data fusion device, including
Classification extraction module 1 is carried out for the collected data information of multisensor syste to electric power wearable device
Classification is extracted, unrelated data information is rejected and acquisition needs to carry out the data information of data fusion；
Bayes predictive model constructs module 2, for by building bayes predictive model to needing to carry out data fusion
Data information be analyzed and processed；
Weighted Fusion module 3, for being believed by the Weighted Fusion algorithm based on variance evaluation sorted data are extracted
Breath is analyzed and processed.
Preferably, the classification extraction module 1 includes:
Taxon, for the collected data information of the multisensor syste to be divided into redundancy, complementary information
And cooperative information；
Culling unit, for rejecting the redundancy and complementary information；
Selection unit, for being analyzed and processed by constructing bayes predictive model to cooperative information.
In the present embodiment, the bayes predictive model building module 2 includes:
Acquiring unit, for obtaining optimal bayesian estimation value；
Comparing unit, for comparing the optimal bayesian estimation value with preset value, when optimal bayesian estimation
When value is greater than preset value, the data are rejected.
Wherein, the acquiring unit includes:
Presumption units, for establishing n Bayesian forecasting moulds to be selected to the collected data information of multisensor syste
Type speculates the probability of sensor failure using each bayes predictive model to be selected；
It is weighted and averaged unit, for by predictions of the n bayes predictive model to be selected for sensor failure
Conclusion is weighted and averaged, and resulting weight is the optimal bayesian estimation value.
Preferably, the Weighted Fusion module 3 includes:
Average calculation unit, for calculating the average value of moment multiple sensor signals；
Average variance computing unit estimates the side of each sensor mean deviation signal for setting data window length T
Difference；
Minimum variance computing unit, for estimating the minimum of each sensor according to the variance of the mean deviation signal
Variance；
Integrated unit, for calculating the collected data information of multisensor syste according to the minimum variance of each sensor
Weighted Fusion value.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (9)
1. a kind of multi-source data fusion method towards electric power wearable device characterized by comprising
S10: classification is extracted to the collected data information of the multisensor syste of electric power wearable device, is rejected unrelated
Data information simultaneously obtains the data information for needing to carry out data fusion；
S20: by building bayes predictive model to needing the data information for carrying out data fusion to be analyzed and processed；
S30: it is analyzed and processed by the Weighted Fusion algorithm based on variance evaluation to sorted data information is extracted.
2. the multi-source data fusion method according to claim 1 towards electric power wearable device, which is characterized in that described
S10 includes:
S101: the collected data information of the multisensor syste is divided into redundancy, complementary information and cooperative information；
S102: the redundancy and complementary information are rejected；
S103: step S20 is executed to the cooperative information.
3. the multi-source data fusion method according to claim 1 towards electric power wearable device, which is characterized in that described
S20 includes:
S201: optimal bayesian estimation value is obtained comprising:
S2011: establishing n bayes predictive models to be selected to the collected data information of multisensor syste, using it is each to
Bayes predictive model is selected to speculate the probability of sensor failure；
S2012: the n bayes predictive models to be selected are weighted the prediction conclusion of sensor failure flat
, resulting weight is the optimal bayesian estimation value；
S202: the optimal bayesian estimation value is compared with preset value, when optimal bayesian estimation value is greater than preset value
When, reject the data.
4. the multi-source data fusion method according to claim 1 towards electric power wearable device, which is characterized in that described
S30 includes:
S301: the average value of t moment multiple sensor signals is calculated
S302: setting data window length T estimates the variance of each sensor mean deviation signal；
S303: the minimum variance of each sensor is estimated according to the variance of the mean deviation signal；
S304: the Weighted Fusion value of the collected data information of multisensor syste is calculated according to the minimum variance of each sensor.
5. a kind of multisource data fusion device towards electric power wearable device characterized by comprising
Classify extraction module, is extracted point for the collected data information of multisensor syste to electric power wearable device
Class, rejects unrelated data information and acquisition needs to carry out the data information of data fusion；
Bayes predictive model constructs module, for by constructing bayes predictive model to the data for needing to carry out data fusion
Information is analyzed and processed；
Weighted Fusion module, for being carried out by the Weighted Fusion algorithm based on variance evaluation to sorted data information is extracted
Analysis processing.
6. the multisource data fusion device according to claim 5 towards electric power wearable device, which is characterized in that described
Classification extraction module include:
Taxon, for the collected data information of the multisensor syste to be divided into redundancy, complementary information and association
Same information；
Culling unit, for rejecting the redundancy and complementary information；
Selection unit, for being analyzed and processed by constructing bayes predictive model to cooperative information.
7. the multisource data fusion device according to claim 5 towards electric power wearable device, which is characterized in that described
Bayes predictive model constructs module
Acquiring unit, for obtaining optimal bayesian estimation value；
Comparing unit, for comparing the optimal bayesian estimation value with preset value, when optimal bayesian estimation value is big
When preset value, the data are rejected.
8. the multisource data fusion device according to claim 7 towards electric power wearable device, which is characterized in that described
Acquiring unit includes:
Presumption units are utilized for establishing n bayes predictive models to be selected to the collected data information of multisensor syste
Each bayes predictive model to be selected speculates the probability of sensor failure；
It is weighted and averaged unit, for the prediction conclusion by the n bayes predictive models to be selected for sensor failure
It is weighted and averaged, resulting weight is the optimal bayesian estimation value.
9. the multisource data fusion device according to claim 5 towards electric power wearable device, which is characterized in that described
Weighted Fusion module includes:
Average calculation unit, for calculating the average value of t moment multiple sensor signals
Average variance computing unit estimates the variance of each sensor mean deviation signal for setting data window length T；
Minimum variance computing unit, for estimating the minimum variance of each sensor according to the variance of the mean deviation signal；
Integrated unit, for calculating the weighting of the collected data information of multisensor syste according to the minimum variance of each sensor
Fusion value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910596096.5A CN110298409A (en) | 2019-07-03 | 2019-07-03 | Multi-source data fusion method towards electric power wearable device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910596096.5A CN110298409A (en) | 2019-07-03 | 2019-07-03 | Multi-source data fusion method towards electric power wearable device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110298409A true CN110298409A (en) | 2019-10-01 |
Family
ID=68030108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910596096.5A Pending CN110298409A (en) | 2019-07-03 | 2019-07-03 | Multi-source data fusion method towards electric power wearable device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110298409A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113555132A (en) * | 2020-04-24 | 2021-10-26 | 华为技术有限公司 | Multi-source data processing method, electronic device and computer-readable storage medium |
| CN113609360A (en) * | 2021-08-19 | 2021-11-05 | 武汉东湖大数据交易中心股份有限公司 | Scene-based multi-source data fusion analysis method and system |
| CN117633706A (en) * | 2023-11-30 | 2024-03-01 | 众悦(威海)信息技术有限公司 | Data processing method for information system data fusion |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105403245A (en) * | 2015-10-16 | 2016-03-16 | 沈阳农业大学 | Sunlight greenhouse wireless sensor multi-data fusion method |
| CN105678337A (en) * | 2016-01-12 | 2016-06-15 | 国网技术学院 | Information fusion method in intelligent transformer station fault diagnosis |
| CN108960334A (en) * | 2018-07-12 | 2018-12-07 | 中国人民解放军陆军炮兵防空兵学院郑州校区 | A kind of multi-sensor data Weighted Fusion method |
| CN109766941A (en) * | 2019-01-02 | 2019-05-17 | 青岛理工大学 | Farmland data fusion processing method based on multiple sensors |
-
2019
- 2019-07-03 CN CN201910596096.5A patent/CN110298409A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105403245A (en) * | 2015-10-16 | 2016-03-16 | 沈阳农业大学 | Sunlight greenhouse wireless sensor multi-data fusion method |
| CN105678337A (en) * | 2016-01-12 | 2016-06-15 | 国网技术学院 | Information fusion method in intelligent transformer station fault diagnosis |
| CN108960334A (en) * | 2018-07-12 | 2018-12-07 | 中国人民解放军陆军炮兵防空兵学院郑州校区 | A kind of multi-sensor data Weighted Fusion method |
| CN109766941A (en) * | 2019-01-02 | 2019-05-17 | 青岛理工大学 | Farmland data fusion processing method based on multiple sensors |
Non-Patent Citations (3)
| Title |
|---|
| SHESHENG GAO.ET AL: ""Random Weighting Method for"", 《IEEE》 * |
| 谢苗苗等: ""基于多源数据融合的鸡舍环境智能监控技术研究"", 《山东农业工程学院学报》 * |
| 陈淑鑫等: ""信息融合多传感器故障贝叶斯可靠性分析"", 《齐齐哈尔大学学报（自然科学版）》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113555132A (en) * | 2020-04-24 | 2021-10-26 | 华为技术有限公司 | Multi-source data processing method, electronic device and computer-readable storage medium |
| CN113609360A (en) * | 2021-08-19 | 2021-11-05 | 武汉东湖大数据交易中心股份有限公司 | Scene-based multi-source data fusion analysis method and system |
| CN117633706A (en) * | 2023-11-30 | 2024-03-01 | 众悦(威海)信息技术有限公司 | Data processing method for information system data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102789545B (en) | Based on the Forecasting Methodology of the turbine engine residual life of degradation model coupling | |
| CN110298409A (en) | Multi-source data fusion method towards electric power wearable device | |
| CN108090515B (en) | Data fusion-based environment grade evaluation method | |
| CN110232524A (en) | Social networks cheats the construction method of model, antifraud method and apparatus | |
| CN108875823B (en) | Evidence combination method based on evidence measurement standard | |
| Jiang et al. | A multisensor cycle-supervised convolutional neural network for anomaly detection on magnetic flux leakage signals | |
| CN110188882A (en) | A kind of high conflicting evidence fusion method based on fuzzy reasoning | |
| CN103631145B (en) | Multiple operating modes process monitoring method and system based on monitor control index switching | |
| CN113569445A (en) | Steel structure health monitoring system and method based on digital twinning technology | |
| CN103473459A (en) | Method of processing and fusing multisystem big data | |
| CN109829136A (en) | A kind of method for predicting residual useful life and system of the degeneration equipment with random jump | |
| CN110175247A (en) | A method of abnormality detection model of the optimization based on deep learning | |
| CN115063017A (en) | Monitoring and evaluating system and method for small and medium-span bridge structure | |
| CN112906288A (en) | Switch cabinet state evaluation method based on normal sample data | |
| CN115526515A (en) | Safety monitoring system of gate for water conservancy and hydropower | |
| CN101751018A (en) | Distributing data monitoring and prealarming system under test environment | |
| CN117171702A (en) | Multi-mode power grid fault detection method and system based on deep learning | |
| CN117349595A (en) | Industrial scene-oriented extremely few sample fault diagnosis method | |
| CN114048546B (en) | Method for predicting residual service life of aeroengine based on graph convolution network and unsupervised domain self-adaption | |
| CN113312696B (en) | Bridge health condition dynamic prediction alarm method based on ARIMA algorithm | |
| CN116957361B (en) | Ship task system health state detection method based on virtual-real combination | |
| CN115754199B (en) | Water quality detection method based on membership function and principal component analysis | |
| Wu et al. | Traffic sensor health monitoring using spatiotemporal graphical modeling | |
| CN117235540A (en) | Sensor dynamic information linkage analysis method based on feature matching fusion | |
| CN103475527A (en) | Network management fault reliability analyzing system and 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191001 |