CN106878375A - A kind of cockpit pollutant monitoring method based on distribution combination sensor network - Google Patents
A kind of cockpit pollutant monitoring method based on distribution combination sensor network Download PDFInfo
- Publication number
- CN106878375A CN106878375A CN201611200530.6A CN201611200530A CN106878375A CN 106878375 A CN106878375 A CN 106878375A CN 201611200530 A CN201611200530 A CN 201611200530A CN 106878375 A CN106878375 A CN 106878375A
- Authority
- CN
- China
- Prior art keywords
- node
- pollutant
- sensor
- specified
- correspondence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Physics & Mathematics (AREA)
- Food Science & Technology (AREA)
- Biochemistry (AREA)
- Combustion & Propulsion (AREA)
- General Physics & Mathematics (AREA)
- Medicinal Chemistry (AREA)
- Pathology (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a kind of cockpit pollutant monitoring method based on distribution combination sensor network, it is used to improve the accuracy and reliability of cockpit pollutant measurement.Methods described includes：Distributed combination sensor network is set up in cockpit to be measured, each node in the distributed combination sensor network is provided with for monitoring same specified pollutant but the different main and auxiliary sensor of measuring principle；To each node initializing；The main and auxiliary sensor of each node is measured and calculates measurement value sensor error respectively；Each node determines the nodal information of itself and propagates to neighbor node；Measurement value sensor error and its information of neighbor nodes to each node carry out the optimal error estimates value that uniformity Kalman filtering obtains each junction sensor；The corresponding master reference measured value of respective nodes is corrected using the optimal error estimates value of sensor, the specified pollutant concentration value of respective nodes is obtained.The method increase the accuracy and reliability of cockpit pollutant monitoring.
Description
Technical field
Aircraft cockpit environmental monitoring of the present invention, more particularly to a kind of cockpit based on distribution combination sensor network
Pollutant monitoring method.
Background technology
The positive pressure cabin of air hermetic ensure that passenger and the crew safe flight in the adverse circumstances of high-altitude.With aviation
Industrial continues to develop, and civil aviaton's each side it is also proposed requirement higher to the environmental quality in cockpit, and various countries and area exist
All clear stipulaties limit value of relevant environmental parameter in corresponding civil aviaton's standard.But these standards just for temperature, pressure and
The basic environment parameter such as humidity, unique regulation pollutant index --- smog is also intended only as a kind of nondominant hand of fire monitoring
Section.Can actual conditions be：The closing of aircraft cockpit narrow space, densely populated place, the pollutant not found in time is by a few hours
Directly infringement is caused in flight time to passenger and crew.The cabin ambient of safety and comfort is passenger's selection airline boat
The key factor of class, is also the healthy necessary guarantee of long-term work crew wherein, and what is particularly newly dispatched from the factory flies
Whether machine disclosure satisfy that passenger's comfort level aboard, can provide a kind of research method for the authorization of the seaworthiness of cabin ambient.
The actual measurement work of correlation, the pollutant in cockpit are all carried out for aircraft cockpit pollutant in recent years both at home and abroad
In monitoring process, due to local single sensor oversensitiveness, under noise jamming there is larger error and often in measurement result
There is false alarm.In practical flight, once there occurs grave warning, pilot operating aircraft must be made preparation for dropping at once, therefore
False alarm will have a strong impact on normal flight plan, cause the loss of a large amount of manpower financial capacities.Can using the method for increasing alarm threshold value
To reduce false alarm rate, but generate serious potential safety hazard.
The content of the invention
The present invention provides a kind of cockpit pollutant monitoring method based on distribution combination sensor network, is used to improve seat
The accuracy and reliability of cabin pollutant measurement.
The present invention provides a kind of cockpit pollutant monitoring method based on distribution combination sensor network, including：
Distributed combination sensor network is set up in cockpit to be measured；Each in the distributed combination sensor network
Node is provided with the sensor for monitoring at least one specified pollutant, and for every kind of specified pollutant, is provided with difference
The master reference of measuring principle and auxiliary sensor；
The neighbor node of each node is defined, and each node i correspondence l kinds of Initialize installation specify the sensor of pollutant
Estimate gain battle array PilAnd state estimation；
For each node, master reference and auxiliary sensor for monitoring same pollutant are measured, worked as respectively
Front nodal point correspondence l kinds specify the master reference measured value Z of pollutantPilWith auxiliary measurement value sensor ZSil；
According to formula Zil=ZPil-ZSilCalculate the measurement value sensor error of each node various specified pollutants of correspondence
Zil；
Each node determines the nodal information of itself and propagates to neighbor node；The nodal information includes present node correspondence
The state estimation of the sensor of various specified pollutants, the information vector and information of the various specified pollutants of present node correspondence
Matrix；
For each node, according to the measurement value sensor error Z of the present node every kind of specified pollutant of correspondenceilWith it is current
The nodal information of all neighbor nodes of node, is filtered by uniformity Kalman filter and obtains the present node every kind of finger of correspondence
Determine the optimal error estimates value of the sensor of pollutant
For each node, the optimal error estimates value of the sensor of pollutant is specified using present node correspondence l kindsCorrection present node correspondence l kinds specify the master reference measured value of pollutant, and the l kinds for obtaining present node specify dirt
Dye thing concentration value：
Wherein, i=1,2 ..., N；N is the nodes in the distributed combination sensor network；L=1,2 ..., L,
L is the specified pollutant kind number of each node monitors.
In one embodiment, each node determines the nodal information of itself, including：
Observation noise covariance matrix R of each node according to the previously given present node various specified pollutants of correspondenceil,
According to formulaWithCalculate the information matrix U of the present node various specified pollutants of correspondencei
With information vector ui；Wherein, HilThe combined error observation matrix of pollutant is specified for known i-th node correspondence l kinds；
Present node is corresponded to each node the state estimation of the sensor of various specified pollutantsPresent node pair
Answer the information vector u of various specified pollutantsiWith information matrix UiIt is defined as the nodal information of present node.
In one embodiment, the measurement value sensor error Z according to the present node every kind of specified pollutant of correspondenceil
With the nodal information of all neighbor nodes of present node, filtered by uniformity Kalman filter and obtain present node correspondence
The optimal error estimates value of the sensor of every kind of specified pollutantIncluding：
Present node receives the nodal information of all neighbor nodes of itself；
Pollutant is specified for the l kinds that present node is monitored, all neighbor nodes correspondence of present node is same
Specifying the information vector and information matrix of pollutant carries out following information fusion：
Present node updates the evaluated error covariance matrix M that present node correspondence l kinds specify the sensor of pollutantil
For：
According to formulaEnter the state estimation of line sensor, obtain
Present node correspondence l kinds specify the state estimation of the sensor of pollutant
According to formulaCalculating present node correspondence l kinds specifies the Optimal error of the sensor of pollutant to estimate
Evaluation
Wherein, j ∈ Ji=Ni∪ { i }, NiIt is i-th neighbor node set of node, variable rilFor：
θ is system communication cycle, and matrix takes Frobenius norms.
In one embodiment, after the l kinds for obtaining present node specify pollutant concentration value, also include：
Judge whether the various specified pollutant concentration value of each node is more than the specified pollution of corresponding species set in advance
Thing concentration threshold；
When any one the specified pollutant concentration value for having node is dense more than the specified pollutant of corresponding species set in advance
During degree threshold value, the exceeded alarm of pollutant concentration is sent.
In one embodiment, preset whether the various specified pollutant concentration value for judging each node is more than
Corresponding species specified pollutant concentration threshold value after, also include：
When the specified pollutant of the various specified pollutant concentration value no more than corresponding species set in advance of each node
During concentration threshold, the air quality index AQI of each node is calculated according to following formulai：
Wherein, ZblFor l kinds set in advance specify pollutant concentration threshold value, αlFor l kinds set in advance specify dirt
Contaminate the index weightses of thing.
In one embodiment, the exceeded air quality index alarmed or calculate each node of pollutant concentration is sent described
Afterwards, also include：
Present node is corresponded to each node the estimation gain battle array P of the sensor of various specified pollutantsilAnd state estimationIt is updated to：Pil=AMilAT,
Return and perform for each node, master reference and auxiliary sensor for monitoring same pollutant are measured respectively
The step of.
In one embodiment, the specified pollutant is carbon dioxide, carbon monoxide, ozone, pellet, third
At least one in ketone, ethanol, formaldehyde, toluene, dichloromethane, endotoxin, microorganism.
Some beneficial effects of the invention can include：
The cockpit pollutant monitoring method based on distribution combination sensor network that the present invention is provided combines distribution
Sensor network is used in being introduced into the monitoring of cockpit pollutant, is provided for measuring same pollutant but not in single measuring node
With the sensor of operation principle, measurement data is merged using uniformity Kalman filtering algorithm, finally obtained accurate
Pollutant concentration value, realizes the accurate measurement of cockpit pollutant, and monitoring result reliability is high.Each junction sensor need not be with institute
There is node to enter row information to exchange, communication is occurred over just between neighbours' sensor node, can effectively reduce data fusion calculating
Complexity, improves the real-time of monitoring, while reducing sensor energy consumption, and avoids reliability short slab in global information fusion
Problem, each junction sensor is combined the globally consistent estimation of information realization of the measured value of itself and neighbor node.
Other features and advantages of the present invention will be illustrated in the following description, also, the partly change from specification
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by the explanations write
Specifically noted structure is realized and obtained in book, claims and accompanying drawing.
Below by drawings and Examples, technical scheme is described in further detail.
Brief description of the drawings
Accompanying drawing is used for providing a further understanding of the present invention, and constitutes a part for specification, with reality of the invention
Applying example is used to explain the present invention together, is not construed as limiting the invention.In the accompanying drawings：
Fig. 1 is the output calibration data fusion schematic diagram of a certain node i in the embodiment of the present invention；
Fig. 2 is a kind of cockpit pollutant monitoring method based on distribution combination sensor network in the embodiment of the present invention
Flow chart；
Fig. 3 is another cockpit pollutant monitoring method based on distributed combination sensor network in the embodiment of the present invention
Flow chart；
Fig. 4 is the flow chart of step S106/S306 in the embodiment of the present invention；
Fig. 5 is another cockpit pollutant monitoring method based on distributed combination sensor network in the embodiment of the present invention
Flow chart；
Fig. 6 is the Digital Simulation schematic diagram in the embodiment of the present invention for node i；
Fig. 7 be the embodiment of the present invention in emulate with 10 cockpit pollutant monitoring network topological diagrams of node；
Fig. 8 is state averaged power spectrum error change figure in a simulation example；
Fig. 9 is state average homogeneity error change figure in a simulation example.
Specific embodiment
To solve problem of the prior art, the present invention uses distributed sensor networks, and is placed in each node region
The sensor of different type pollutant is monitored, for a certain pollutant classification, the sensor of different measuring principles is chosen, by these
The measurement result of sensor carries out data fusion to make up the deficiency of independent type sensor.It is well known that in sensor network
Each sensor can obtain close but not quite identical measurement result under the effect of numerous enchancement factors.Traditional solution
It is that a data fusion center is set, fusion is weighted to the data that all the sensors are monitored.But program data communication
Amount is big, and data fusion center will cause whole system failure once failing, and reliability is poor.And cockpit pollutant monitoring is needed
High real-time and accuracy are wanted, the appearance of pollutant can in time be monitored by monitoring network and the pollutant is judged
Whether burst size meets standard.Therefore the present invention uses the distributed consensus Kalman filter to carry out data fusion to reach standard
Really measure and improve the reliability of measurement result.
Cockpit pollutant monitoring method proposed by the present invention choose cockpit to be measured in exemplary position as network node, including
The place sensors such as cockpit top, bottom, window, seat, constitute distributed combination sensor network, and each node is set
There is the sensor for monitoring at least one specified pollutant, and for every kind of specified pollutant, be provided with different measuring principles
Master reference and auxiliary sensor.The topology diagram that G=(V, E, A) is the distributed combination sensor network selected is defined,
Wherein V={ v1,v2,...vnIt is sensor node set；For sensor node information exchanges even line set.I-th
The neighbor node collection of individual node shares Ni={ vj∈V|(vi,vj) ∈ E represent, and the number of its neighbor node is referred to as the node
Degree and be designated as di=| Ni|.The degree matrix D of topological diagram G is defined as the diagonal matrix with each sensor node degree as diagonal element, i.e.,
D=diag { d1,...dN}.A=[aij] it is the adjacency matrix for representing each sensor correspondence, when i-th sensor and jth
When individual sensor is communicated, aijValue is 1, is otherwise 0.The LaPlacian matrix definition for scheming G is L=D-A, if matrix L
Containing a nonzero eigenvalue, then illustrate that undirected topological diagram G is UNICOM.
When the measurement value sensor of each node to distribution combination sensor network carries out data fusion, using figure
Output calibration data fusion structure shown in 1, uniformity Kalman filter measures the main and auxiliary of same pollutant to node i
The measurement result Z of sensorPiAnd ZSiDifference carry out data fusion, uniformity Kalman filter exports the two sensors mistake
Difference ZiOptimal State EstimationOutput calibration is carried out to the measured value of master reference with this error estimate again, reality is obtained
The optimal estimation value of pollutant concentration.
In the Data Fusion Structure shown in Fig. 1, uniformity Kalman filter is entered to the measurement result of main and auxiliary sensor
Row data fusion, therefore firstly the need of the error dynamics equation of the combination sensor for setting up the two sensor combinations.According to being
System actual performance statistical analysis and the error analysis of sensor, can obtain the error model of each sensor, for mark is simple, adopt
Updated with the subsequent time of subscript+represent variable states.
Shown in master reference error dynamics equation such as formula (1)：
Shown in auxiliary sensor error dynamical equation such as formula (2)：
Wherein,It is the state vector of main and auxiliary sensor；ZPi∈Rp,ZSi∈RsIt is main and auxiliary sensing
The output vector of device；(AP,BP,HPi) and (AS,BS,HSi) be appropriate dimension parameter matrix.wP,vPi,wS,vSiFor it is one-dimensional mutually
Independent white Gaussian noise signal, meets：
E[w(k)w(l)T]=Q (k) δkl
E[vi(k)vj(l)T]=Rij(k)δkl
Wherein, δklIt is unit impulse function, i.e., value is 1 as k=l, is otherwise 0；Q and RijFor corresponding noise is assisted
Variance matrix, for mark is convenient, by RijIt is abbreviated as Ri。
Actual performance and error analysis according to system, can be obtained the global error side of combination sensor by formula (1), (2)
Journey, in this, as the system equation of uniformity Kalman filter.Because the sampling instant of main and auxiliary sensor is it cannot be guaranteed that completely
Unanimously, in order to simplify the systematic observation equation at alignment moment, the measurement noise of auxiliary sensor can be ignored, and by master reference
Various errors are classified as the measurement noise of combination sensor.The error equation for obtaining combination sensor is as follows：
For convenience of hereafter analyzing, formula (3) is expressed as Unified Form：
Contrast (3) and formula (4), obtains combined error state vector for x=col (xP,xS), combined error system noise is w
=col (wP,wS)；Assuming that combined error observation noise takes master reference observation noise, i.e. vi=vPi.Combined error system equation
Parameter is A=diag (AP,AS)；B=diag (BP,BS)；HiIt is i-th pollution for present combination Sensor monitoring of node
The combined error observation matrix of thing, can actually obtain according to sensor measurement errors principle and engineering.
For the distributed system represented by formula (4), the Kalman filter structure of its discrete form is as follows.
Wherein KiAnd CiRespectively filtering gain matrix and consistency matrix.WithIt is respectively the estimate of tested state x
With predicted value, it is as follows that it counts implication：
Wherein Z (k)=col { Z1(k),...,ZN(k) } be each sensor observation composition column vector.Shape is defined respectively
State evaluated error eiAnd status predication errorIt is as follows：
Then there is MijAnd PijEvaluated error covariance matrix and predicting covariance battle array are represented respectively, and have following statistics to contain
Justice：
Mij(k)=E [ηi(k)ηj(k)T] (10)
In order to mark simplicity, as node i=j, respectively by MijAnd PijIt is abbreviated as MiAnd Pi.Define weighted measuresAnd information matrixThen message form can will be expressed as with upper filter：
Uniformity Kalman's information filter, use state predicting covariance battle array P are referred to as with upper filteriAnd state
Evaluated error covariance matrix MiTo correct status predication valueTo obtain the estimation to error stateIts data fusion effect is with good expansibility, and goes for large-scale distributed monitoring net
Network.Be used for uniformity Kalman information filter in proposed distributed combination sensor network by the present invention.
The preferred embodiments of the present invention are illustrated below in conjunction with accompanying drawing, it will be appreciated that preferred reality described herein
Apply example to be merely to illustrate and explain the present invention, be not intended to limit the present invention.
Fig. 2 is a kind of cockpit pollutant monitoring method based on distribution combination sensor network in the embodiment of the present invention
Flow chart.As shown in Fig. 2 the method comprises the following steps S101-S107：
S101：Distributed combination sensor network is set up in cockpit to be measured；In the distributed combination sensor network
Each node be provided with sensor for monitoring at least one specified pollutant, and for every kind of specified pollutant, set
There are the master reference and auxiliary sensor of different measuring principles.
Below for convenience of description, l kinds any two node monitored respectively specify pollutant to be set as same type
Pollutant.If for example, the distributed combination sensor network has 2 nodes, each node is provided with for monitoring two
Carbonoxide, carbon monoxide, ozone totally 3 kinds of sensors of specified pollutant, the 1st kind, the 2nd kind, the 3rd kind of finger that node 1 is monitored
Determine pollutant and be appointed as carbon dioxide, carbon monoxide, ozone successively, then the 1st kind, the 2nd kind, the 3rd kind for monitoring node 2 is specified
Pollutant is also appointed as carbon dioxide, carbon monoxide, ozone successively.
S102：The neighbor node of each node is defined, and each node i correspondence l kinds of Initialize installation specify the biography of pollutant
The estimation gain battle array P of sensorilAnd state estimationFor：
Pil=Pil(0),
Wherein, i=1,2 ..., N；N is the nodes in the distributed combination sensor network；L=1,2 ..., L,
L is the specified pollutant kind number of each node monitors, i.e.,：Subscript i represents node coefficient, and l to be represented and specify pollutant with l kinds
Corresponding parameter.
S103：For each node i, for monitoring same pollutant but the different master reference of measuring principle and auxiliary biography
Sensor is measured respectively, obtains the master reference measured value Z that present node correspondence l kinds specify pollutantPilMeasured with auxiliary sensor
Value ZSil。
S104：Calculate the measurement value sensor error Z of each node various specified pollutants of correspondenceil, computing formula is：
Zil=ZPil-ZSil。
S105：Each node determine itself nodal information (uil, Uil) and propagate to neighbor node.Wherein,It is to work as
The state estimation of the sensor of the various specified pollutants of front nodal point correspondence, uilIt is the present node various specified pollutants of correspondence
Information vector, UilIt is the information matrix of the present node various specified pollutants of correspondence.
Wherein, each node is first according to the observation noise association of the previously given present node various specified pollutants of correspondence
Variance matrix Ril, according to formulaWithCalculate the present node various specified pollutants of correspondence
Information matrix UilWith information vector uil；Present node is corresponded to subsequent each node the shape of the sensor of various specified pollutants
State estimateThe information vector u of the various specified pollutants of present node correspondenceilWith information matrix UilIt is defined as present node
Nodal information (uil, Uil)；Wherein, HilFor known i-th node correspondence l kinds specify the combined error of pollutant to see
Measured value matrix.
S106：For each node, according to the measurement value sensor error Z of the present node every kind of specified pollutant of correspondenceil
With the nodal information of all neighbor nodes of present node, filtered by uniformity Kalman filter and obtain present node correspondence
The optimal error estimates value of the sensor of every kind of specified pollutant
S107：For each node, the Optimal error of the sensor of pollutant is specified to estimate using present node correspondence l kinds
EvaluationCorrection present node correspondence l kinds specify the master reference measured value of pollutant, and the l kinds for obtaining present node are specified
Pollutant concentration value ZOil。
Wherein, updating formula is：ZOilFor the l kinds of node i specify pollutant concentration value, ZPilFor
Node i correspondence l kinds specify the master reference measured value of pollutant,The sensor of pollutant is specified for node i correspondence l kinds
Optimal error estimates value.
Technical scheme provided in an embodiment of the present invention, distribution combination sensor network is introduced into the monitoring of cockpit pollutant
Use, the sensor for measuring same pollutant but different operating principle is provided in single measuring node, using uniformity
Kalman filtering algorithm is merged to measurement data, finally obtains accurate pollutant concentration value, realizes cockpit pollutant
Accurate measurement, monitoring result reliability is high.
Fig. 3 is another cockpit pollutant monitoring method based on distributed combination sensor network in the embodiment of the present invention
Flow chart.As shown in figure 3, the method comprises the following steps S201-S210：
S201：Distributed combination sensor network is set up in cockpit to be measured.
S202：The neighbor node of each node is defined, and each node i correspondence l kinds of Initialize installation specify the biography of pollutant
The estimation gain battle array P of sensorilAnd state estimation
S203：For each node i, for monitoring same pollutant but the different master reference of measuring principle and auxiliary biography
Sensor is measured respectively, obtains the master reference measured value Z that present node correspondence l kinds specify pollutantPilMeasured with auxiliary sensor
Value ZSil。
S204：Calculate the measurement value sensor error Z of each node various specified pollutants of correspondenceil。
S205：Each node determines the nodal information of itself and propagates to neighbor node.
S206：For each node, according to the measurement value sensor error Z of the present node every kind of specified pollutant of correspondenceil
With the nodal information of all neighbor nodes of present node, filtered by uniformity Kalman filter and obtain present node correspondence
The optimal error estimates value of the sensor of every kind of specified pollutant
S207：For each node, the Optimal error of the sensor of pollutant is specified to estimate using present node correspondence l kinds
EvaluationCorrection present node correspondence l kinds specify the master reference measured value of pollutant, and the l kinds for obtaining present node are specified
Pollutant concentration value ZOil。
In the present embodiment, the step S101-S107 classes in the specific implementation method and above-described embodiment of step S201-S207
Seemingly, here is omitted.
S208：Judge whether the various specified pollutant concentration value of each node is specified more than corresponding species set in advance
Pollutant concentration threshold value；When specified dirt of any one the specified pollutant concentration value for having node more than corresponding species set in advance
During dye thing concentration threshold, step S209 is performed；Otherwise, i.e., when the various specified pollutant concentration value of each node is no more than advance
During the specified pollutant concentration threshold value of the corresponding species of setting, step S210 is performed.
Because pollutant concentration monitoring is not also included investigation scope by current civil aviaton's air worthiness regulation and design standard.The present invention
With reference to existing measured data and Indoor environment environmental standard, with reference to the characteristic of aircraft cockpit restricted clearance, it is considered to dirty below
Not, related pollutant concentration threshold value and measuring method is as shown in table 1 for dye species：
The cockpit pollutant monitoring classification of table 1
Preferably, the specified pollutant of each node monitors is carbon dioxide, carbon monoxide, ozone, pellet, third
At least one in ketone, ethanol, formaldehyde, toluene, dichloromethane, endotoxin, microorganism.
S209：Send the exceeded alarm of pollutant concentration.
In this step, the exceeded node identification of pollutant concentration and exceeded pollution species can be provided in alarm
Type.
S210：Calculate the air quality index AQI of each nodei：
Wherein, ZblFor l kinds set in advance specify pollutant concentration threshold value, αlFor l kinds set in advance specify dirt
Contaminate the index weightses of thing.
In the present embodiment, measured by distribution combination sensor network and uniformity Kalman filter carries out data and melts
After the pollutant concentration that merging obtains to node, can also judge that each monitoring is saved by the pollutant concentration threshold value that pre-sets
Whether the various pollutant concentrations of point are exceeded, and remind user in exceeded alarm body, additionally, give Cabin air quality commenting
Valency index calculation method, calculates the air quality index of each node when not exceeded, to facilitate user to each key position of cockpit
Air quality carry out monitor in real time.
In one embodiment, as shown in figure 4, the filtering method of the uniformity Kalman filter of step S106/S206
Following steps S301-S305 can be embodied as：
S301：Each node i receive all neighbor node j of itself nodal information (ujl, Ujl)。
S302：Pollutant is specified for the l kinds that present node i is monitored, by all neighbor nodes j pairs of present node i
Answer the information vector u of same specified pollutantjlWith information matrix UjlInformation fusion is carried out according to equation below (14)：
Wherein, j ∈ Ji=Ni∪ { i }, NiIt is i-th neighbor node set of node.
S303：Present node i updates the sensor of the present node i specified pollutants of correspondence l kinds according to formula (15)
Evaluated error covariance matrix MilFor：
S304：Enter the state estimation of line sensor according to formula (16), obtain present node i correspondence l kinds and specify pollution
The state estimation of the sensor of thing
S305：Calculating present node i correspondence l kinds according to formula (17) specifies the Optimal error of the sensor of pollutant to estimate
Evaluation
Wherein, variable rilFor：
θ is system communication cycle, and matrix takes Frobenius norms.
Fig. 5 is another cockpit pollutant monitoring method based on distributed combination sensor network in the embodiment of the present invention
Flow chart.As shown in figure 5, the method comprises the following steps S401-S411：
S401：Distributed combination sensor network is set up in cockpit to be measured.
S402：The neighbor node of each node is defined, and each node i correspondence l kinds of Initialize installation specify the biography of pollutant
The estimation gain battle array P of sensorilAnd state estimation
S403：For each node i, master reference and auxiliary sensor for monitoring same pollutant are measured respectively,
Obtain the master reference measured value Z that present node correspondence l kinds specify pollutantPilWith auxiliary measurement value sensor ZSil。
S404：Calculate the measurement value sensor error Z of each node various specified pollutants of correspondenceil。
S405：Each node determines the nodal information of itself and propagates to neighbor node.
S406：For each node, according to the measurement value sensor error Z of the present node every kind of specified pollutant of correspondenceil
With the nodal information of all neighbor nodes of present node, filtered by uniformity Kalman filter and obtain present node correspondence
The optimal error estimates value of the sensor of every kind of specified pollutantIn the present embodiment, this step uses method reality shown in Fig. 4
It is existing, specifically repeat no more.
S407：For each node, the Optimal error of the sensor of pollutant is specified to estimate using present node correspondence l kinds
EvaluationCorrection present node correspondence l kinds specify the master reference measured value of pollutant, and the l kinds for obtaining present node are specified
Pollutant concentration value ZOil。
S408：Judge whether the various specified pollutant concentration value of each node is specified more than corresponding species set in advance
Pollutant concentration threshold value；When specified dirt of any one the specified pollutant concentration value for having node more than corresponding species set in advance
During dye thing concentration threshold, step S409 is performed；Otherwise, i.e., when the various specified pollutant concentration value of each node is no more than advance
During the specified pollutant concentration threshold value of the corresponding species of setting, step S410 is performed.
S409：The exceeded alarm of pollutant concentration is sent, step S411 is then performed.
S410：Calculate the air quality index AQI of each nodei, then perform step S411.
In the present embodiment, the step S301-S310 classes in the specific implementation method and above-described embodiment of step S401-S410
Seemingly, here is omitted.
S411：Present node is corresponded to each node the estimation gain battle array P of the sensor of various specified pollutantsilAnd state is estimated
EvaluationUpdated according to formula (19), be then returned to perform S403.
Wherein,The state estimation of the sensor of pollutant is specified for present node i correspondence l kinds, by step S406
Obtain.
The present embodiment carries out data and melts by distribution combination sensor network measurement and uniformity Kalman filter
After the pollutant concentration that merging obtains to node, each node upgrades in time the estimation gain battle array P of sensorilAnd state estimationSo as to can return to recycle this method, the monitor in real time to cockpit pollutant concentration is realized.
In the distributed combination sensor network that the present invention is provided, each section is estimated using uniformity Kalman filtering algorithm
The measurement error of the main and auxiliary sensor of point, and for correcting master reference measured value to obtain final pollutant measurement value.Therefore,
The stability of the uniformity Kalman filtering algorithm determines the stability of whole monitoring method.It is local under system noise effect
The state estimation of Kalman filter can deviate actual value, and observation noise can cause each node state estimate can not be complete
Reach unanimity.Method provided in an embodiment of the present invention uniformity Kalman under system noise and observation noise interference is given below
The analysis of filtering algorithm stability.Illustrated infra for convenient, if distributed combination sensor network only monitors a kind of pollution
Thing, then all parameters in the above method for representing that the subscript l of pollutant type can not be considered further that.
In the classical uniformity Kalman filtering algorithm represented by formula (5), by intermediate variable MiSubstitute into Pi +And willSubstitute intoThe filtering algorithm of following more compact form can be obtained.
Before stability analysis is carried out to the uniformity Kalman filtering algorithm represented by formula (20), be first given and be directed to
Two hypotheses of the algorithm parameter, while providing three lemma related to the algorithm.
Assuming that 1：Sytem matrix A is nonsingular all the time in distributed system represented by formula (4).
Assuming that 2：There is arithmetic number q,r,p,So that algorithm parameter matrix has such as lower boundary：
Wherein matrix I is unit battle array.It should be noted that this two hypotheses are not harsh.For by continuous system
The discrete distributed system represented by formula (4) that sampling is obtained, it is assumed that 1 sets up all the time.According to document " Andersen B and
Moore J.Detectability and stabilisability of time-varying discrete-time
linear systems[J].Siam Journal on Control&Optimization.1981,19,(1):20-32.doi:
10.1137/0319002. ", if the systems compliant represented by formula (4) is considerable, PiWith bound.
Lemma 1：For a random process Vk(ξk), if there is real numberAnd 0 ＜ α≤1 so that
E{Vk(ξk)|ξk-1}≤(1-α)Vk-1(ξk-1)+μ (23)
Set up simultaneously, then the square Bounded Index of the random process, i.e.,
And the random process is according to the bounded of probability 1.
In ensuing analysis, lemma 1 be used to judge the stabilization of uniformity Kalman filtering algorithm state estimation
Property.For each measuring node, its parametric boundary condition is as described in lemma 2 and lemma 3：
Lemma 2：Assuming that on the premise of 1 and hypothesis 2, to each node, there is the ＜ κ of real number 0i＜ 1 (i=1 ..., N),
So that
(A-AKiHi)T(Pi +)-1(A-AKiHi)≤(1-κi)Pi -1 (24)
Wherein
Lemma 3：Assuming that on the premise of 1 and hypothesis 2, to each node, there is real number εi＞ 0 (i=1 ..., N), makes
Consider that the Main Conclusions of algorithm stability analysis during noise is as follows：
Theorem 1：The radio sensing network that consideration is made up of N number of node, to the distributed discrete time-varying represented by formula (4)
Uniformity Kalman filtering algorithm represented by system application formula (5).Assuming that on the premise of 1 and 2, if initial predicted is missed
Difference bounded, the then square Bounded Index of the predicated error of algorithm and according to the bounded of probability 1.
Theorem 1 is proved：
According to (8) and (11) are defined, predicated error vector is defined firstAnd corresponding block diagonal battle array P
=diag { P1,...PN}.Following liapunov function is built as the random process in lemma 1：
According in hypothesis 2Can obtain
Wherein,Andp=minp 1,...,p N}.Inequality (28) meets first in lemma 1
Condition, in order to prove the square Bounded Indexes of random process e, it is necessary to consider the mathematical expectation of subsequent time liapunov function
E{V+(e+)}.According to predicated error eiDefinition, can obtain following on eiDynamical equation：
Substituted into the definition (21) of liapunov function, then had
Wherein
Above formula both sides are taken with mathematic expectaion, and apply lemma 2 and lemma 3, formula (30) can abbreviation be
The 3rd and the 4th, analysis above formula right side for convenience, we are by uniformity gain CiIt is defined as form
Λi=(A-AKiHi)T(Pi +)-1(A-AKiHi) (32)
Ci=σ [(A-AKiHi)T(Pi +)-1A]-1Λi(33) (A-AK is multiplied in the right side simultaneously in formula (33) both sidesiHi)T(Pi +)-1A,
Can conveniently obtain
Then (32) will be defined and will substitute into above formula, can obtained by transposition
Ci=σ A-1(A-AKiHi) (35)
(A-AKiHi)-1ACi=σ (36)
Continue for formula (36) to substitute into formula (34), have
According to formula (34) and formula (37), the 3rd and the 4th, formula (31) right side can be expressed as
Define Λ=diag { Λ1,...ΛNAnd q=col { q1,...qN}.The Laplacian Matrix of radio sensing network
Represented with L as previously described, then q can be expressed asWhereinTherefore, formula (38) can further abbreviation
For
Wherein, λminAnd λmaxThe minimum and maximum characteristic value of difference representing matrix.Composite type (31) and formula (39), Wo Menyou
Whereinκ=min { κ1,...κN}.As described in lemma 2, obviously set up with lower inequality
During above formula substituted into formula (40), have
In order to meet the 2nd condition in lemma 1, we must assure that lower inequality such as is set up
Formula (43) can regard the One- place 2-th Order inequality on σ as, and as σ ＜ σ*When the inequality set up, wherein
Simultaneously according to lemma 3, inequality (44) is obviously set up.Therefore, predicated error eiAccording to the square Bounded Index of probability 1.
The embodiment of the present invention is also carried out to the above-mentioned cockpit pollutant monitoring method based on distribution combination sensor network
Emulation, verifying its implementation result.
Fig. 6 show the Digital Simulation schematic diagram of node i, for convenience of description, distributed combination sensor residing for the node
In network, each node is only provided with master reference simulation model and auxiliary sensor Simulation mould for monitoring a kind of pollutant concentration
Type.As shown in fig. 6, the main and auxiliary sensor Simulation model of node i is in respective system noise WiWith measurement noise ViIn the presence of, point
Not Shu Chu pollutant concentration measured value ZPi,ZSi, the difference between the two is measurement error Zi.Uniformity Kalman filter is simultaneously
Also the measurement error Z of the neighbor node j of receiving node ij, output estimation value after being filtered through uniformity Kalman filterWith school
The master reference measured value Z of positive node iPiSo as to obtain final measured value ZOi.Emulation experiment is carried out using DSMC
A large amount of independent repeated trials, are analyzed using the parametric statistics average at each moment to the performance of monitoring network.What is defined is as follows
Performance indications：
State averaged power spectrum error (Mean Estimation Error, MEE)：
ei,k=ZO,k-ZOi,k (47)
State average homogeneity error (Mean Consensus Error, MCE)：
Wherein, subscript k represents the simulation run moment, and N is node total number.
The Mathematical Modeling of main and auxiliary sensor is respective governing equation in Fig. 6, both can come from the reason of measurement error
System Discrimination can be carried out by analysis by actual measurement data again to obtain, it is considered to the amount obtained after system noise and measurement noise
Surveying data will be more nearly reality.It should be noted that limited by System Identification Accuracy and embedded device performance etc., point
The system equation and measurement equation that cloth uniformity Kalman filter is taken can only be to the approximate of actual sensor combination.
Next a simulation example of the invention is given.Deployment has 10 sensor detection networks of node, network
Topological diagram is as shown in fig. 7, wherein each node is only provided with master reference and auxiliary sensing for monitoring a kind of pollutant concentration
Device.Its Laplacian Matrix is：
Consider combination sensor Measuring error model be：
Error original state takes xi(0)=(8,12)T, system noise and measurement noise take covariance for 10 and 100i respectively
Separate white Gaussian noise interference, wherein i be node ID, the different measurement error of each node is embodied with this.System is adopted
The sample cycle is 10ms, and Kalman filter initial prediction error matrix takes Pi(0)=10I2。
Stability analysis is carried out to the system, two hypothesis put forward in above stability analysis are first verified that.For vacation
If 1, each moment AkObviously it is nonsingular.On assume 2, for the system, below inequality set up：
According to formula (25), can calculateκ=0.4515, meet the ＜ of precondition 0 of lemma 2κ＜ 1.According toDetermine
Justice, can calculateAgain by formula (45), σ is obtained*=0.7417 determining taking for parameter of consistency
Value scope.
In current cockpit pollutant monitoring, each sensor uses local Kalman filtering algorithm independent process data.To test
The validity of method provided by the present invention is demonstrate,proved, while entering to local Kalman filtering algorithm and uniformity Kalman filtering algorithm
After 1000 independent Monte-Carlo Simulations of row, the state averaged power spectrum error change of statistics formula (46) is as shown in figure 8, statistics formula
(48) state average homogeneity error is as shown in Figure 9.
Local Kalman filtering algorithm is not due to accounting for the information from surroundings nodes, therefore state estimation general effect
It is poor：Fig. 8 shows its state estimation error higher than the uniformity Kalman filtering algorithm that is carried of the present invention, at the same convergence rate compared with
Slowly；Importantly, the conformity error of local Kalman filtering algorithm is far above the uniformity karr that the present invention is carried in Fig. 9
Graceful filtering algorithm, illustrates that each sensor is larger to the estimate gap of same state, this also explains why single-sensor is easy
Generation false alarm.Because the method that the present invention is provided considers system noise during pollutant monitoring and measures noise, because
This each node estimate can not accurately restrain actual value completely, while the estimate between each node there is also certain error,
But these errors are converged in finite value quickly.The traditional local Kalman filtering algorithm of contrast, institute's extracting method state of the present invention
Estimated accuracy is higher, and each measurement value sensor reaches unanimity, and efficiently solves local single-sensor and triggers because error is larger
The problem of false alarm.
Obviously, those skilled in the art can carry out various changes and modification without deviating from essence of the invention to the present invention
God and scope.So, if these modifications of the invention and modification belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprising these changes and modification.
Claims (7)
1. a kind of based on the distributed cockpit pollutant monitoring method for combining sensor network, it is characterised in that including：
Distributed combination sensor network is set up in cockpit to be measured；Each node in the distributed combination sensor network
The sensor for monitoring at least one specified pollutant is provided with, and for every kind of specified pollutant, is provided with different measurements
The master reference of principle and auxiliary sensor；
The neighbor node of each node is defined, and each node i correspondence l kinds of Initialize installation specify the estimation of the sensor of pollutant
Gain battle array PilAnd state estimation
For each node, master reference and auxiliary sensor for monitoring same pollutant measure, obtain working as prosthomere respectively
Point correspondence l kinds specify the master reference measured value Z of pollutantPilWith auxiliary measurement value sensor ZSil；
According to formula Zil=ZPil-ZSilCalculate the measurement value sensor error Z of each node various specified pollutants of correspondenceil；
Each node determines the nodal information of itself and propagates to neighbor node；The nodal information includes that present node correspondence is various
Specify state estimation, the information vector of the various specified pollutants of present node correspondence and the information square of the sensor of pollutant
Battle array；
For each node, according to the measurement value sensor error Z of the present node every kind of specified pollutant of correspondenceilAnd present node
All neighbor nodes nodal information, filtered by uniformity Kalman filter and obtain the present node every kind of specified dirt of correspondence
Contaminate the optimal error estimates value of the sensor of thing
For each node, the optimal error estimates value of the sensor of pollutant is specified using present node correspondence l kindsSchool
Proper front nodal point correspondence l kinds specify the master reference measured value of pollutant, and the l kinds for obtaining present node specify pollutant dense
Angle value：
Wherein, i=1,2 ..., N；N is the nodes in the distributed combination sensor network；L=1,2 ..., L, L be
The specified pollutant kind number of each node monitors.
2. as claimed in claim 1 based on the distributed cockpit pollutant monitoring method for combining sensor network, its feature exists
In, each node determines the nodal information of itself, including：
Observation noise covariance matrix R of each node according to the previously given present node various specified pollutants of correspondenceil, according to
FormulaWithCalculate the information matrix U of the present node various specified pollutants of correspondenceilWith
Information vector uil；Wherein, HilThe combined error observation matrix of pollutant is specified for known i-th node correspondence l kinds；
Present node is corresponded to each node the state estimation of the sensor of various specified pollutantsPresent node correspondence is each
Plant the information vector u of specified pollutantilWith information matrix UilIt is defined as the nodal information of present node.
3. as claimed in claim 2 based on the distributed cockpit pollutant monitoring method for combining sensor network, its feature exists
In the measurement value sensor error Z according to the present node every kind of specified pollutant of correspondenceilWith all neighbours of present node
The nodal information of node, the sensing for obtaining the present node every kind of specified pollutant of correspondence is filtered by uniformity Kalman filter
The optimal error estimates value of deviceIncluding：
Present node receives the nodal information of all neighbor nodes of itself；
Pollutant is specified for the l kinds that present node is monitored, all neighbor nodes correspondence of present node is same specified
The information vector and information matrix of pollutant carry out following information fusion：
Present node updates the evaluated error covariance matrix M that present node correspondence l kinds specify the sensor of pollutantilFor：
According to formulaEnter the state estimation of line sensor, obtain current
Node correspondence l kinds specify the state estimation of the sensor of pollutant
According to formulaCalculate the optimal error estimates value that present node correspondence l kinds specify the sensor of pollutant
Wherein, j ∈ Ji=Ni∪ { i }, NiIt is i-th neighbor node set of node, variable rilFor：
θ is system communication cycle, and matrix takes Frobenius norms.
4. as described in any one of claims 1 to 3 based on distribution combination sensor network cockpit pollutant monitoring method,
Characterized in that, after the l kinds for obtaining present node specify pollutant concentration value, also including：
Judge whether the various specified pollutant concentration value of each node is dense more than the specified pollutant of corresponding species set in advance
Degree threshold value；
When specified pollutant concentration threshold of any one the specified pollutant concentration value for having node more than corresponding species set in advance
During value, the exceeded alarm of pollutant concentration is sent.
5. as claimed in claim 4 based on the distributed cockpit pollutant monitoring method for combining sensor network, its feature exists
In, the various specified pollutant concentration value for judging each node whether more than corresponding species set in advance specified pollution
After thing concentration threshold, also include：
When the specified pollutant concentration of the various specified pollutant concentration value no more than corresponding species set in advance of each node
During threshold value, the air quality index AQI of each node is calculated according to following formulai：
Wherein, ZblFor l kinds set in advance specify pollutant concentration threshold value, αlFor l kinds set in advance specify pollutant
Index weightses.
6. as claimed in claim 5 based on the distributed cockpit pollutant monitoring method for combining sensor network, its feature exists
In, it is described send pollutant concentration it is exceeded alarm or calculate each node air quality index after, also include：
Present node is corresponded to each node the estimation gain battle array P of the sensor of various specified pollutantsilAnd state estimationMore
It is newly：Pil=AMilAT,
Return to the step for performing and being measured respectively for each node, master reference and auxiliary sensor for monitoring same pollutant
Suddenly.
7. the cockpit pollutant monitoring method based on distributed combination sensor network as claimed in claim 1, described to specify
Pollutant is carbon dioxide, carbon monoxide, ozone, pellet, acetone, ethanol, formaldehyde, toluene, dichloromethane, endogenous toxic material
At least one in element, microorganism.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611200530.6A CN106878375B (en) | 2016-12-22 | 2016-12-22 | A kind of cockpit pollutant monitoring method based on distributed combination sensor network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611200530.6A CN106878375B (en) | 2016-12-22 | 2016-12-22 | A kind of cockpit pollutant monitoring method based on distributed combination sensor network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106878375A true CN106878375A (en) | 2017-06-20 |
| CN106878375B CN106878375B (en) | 2019-06-07 |
Family
ID=59163907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611200530.6A Active CN106878375B (en) | 2016-12-22 | 2016-12-22 | A kind of cockpit pollutant monitoring method based on distributed combination sensor network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106878375B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108593557A (en) * | 2018-03-13 | 2018-09-28 | 杭州电子科技大学 | Based on TE-ANN-AWF mobile pollution source telemetry errors compensation methodes |
| CN109061068A (en) * | 2018-08-15 | 2018-12-21 | 中国民航大学 | The fault-tolerant measurement estimation method of cabin pollutant concentration |
| CN109459527A (en) * | 2018-09-25 | 2019-03-12 | 中国商用飞机有限责任公司 | A kind of monitoring method and monitoring system of aircraft cockpit carbonomonoxide concentration |
| CN109782269A (en) * | 2018-12-26 | 2019-05-21 | 北京壹氢科技有限公司 | A kind of distribution multi-platform cooperative active target tracking |
| CN110312225A (en) * | 2019-07-30 | 2019-10-08 | 平顶山学院 | A kind of wireless sensor hardware device |
| CN111458471A (en) * | 2019-12-19 | 2020-07-28 | 中国科学院合肥物质科学研究院 | Water area detection early warning method based on graph neural network |
| CN111601269A (en) * | 2020-05-15 | 2020-08-28 | 中国民航大学 | Event trigger Kalman consistency filtering method based on information freshness judgment |
| CN112557996A (en) * | 2019-09-26 | 2021-03-26 | 武汉国测数据技术有限公司 | Electric energy measuring system convenient for error checking and error checking method |
| CN114152791A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Three-meter method three-phase electric energy meter structure for user self-checking error and checking method |
| CN114152809A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Intelligent electric meter with error self-checking function and checking method thereof |
| CN114152811A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Electric energy meter with three-way array structure, and constituent measurement system and measurement method thereof |
| CN114152806A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Electric energy sensor with three-way array structure and measurement system and method formed by electric energy sensor |
| CN114152810A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Three-phase electric energy sensor with three-way array structure and measuring system and method thereof |
| CN114152808A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Intelligent electric meter with error self-checking function and checking method thereof |
| CN117241281A (en) * | 2023-11-13 | 2023-12-15 | 中赣通信(集团)有限公司 | Indoor distributed monitoring method and monitoring network |
| CN117713750A (en) * | 2023-12-14 | 2024-03-15 | 河海大学 | Consistency Kalman filtering state estimation method based on fractional power |
| CN117895920A (en) * | 2024-03-13 | 2024-04-16 | 南京工业大学 | Distributed consistency Kalman filtering method for sensor network under communication link fault |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1910428A (en) * | 2003-12-05 | 2007-02-07 | 霍尼韦尔国际公司 | System and method for using multiple aiding sensors in a deeply integrated navigation system |
| CN101505532A (en) * | 2009-03-12 | 2009-08-12 | 华南理工大学 | Wireless sensor network target tracking method based on distributed processing |
| CN102221365A (en) * | 2010-04-19 | 2011-10-19 | 霍尼韦尔国际公司 | Systems and methods for determining inertial navigation system faults |
| CN103063212A (en) * | 2013-01-04 | 2013-04-24 | 哈尔滨工程大学 | Integrated navigation method based on non-linear mapping self-adaptive hybrid Kalman/H infinite filters |
| CN103648108A (en) * | 2013-11-29 | 2014-03-19 | 中国人民解放军海军航空工程学院 | Sensor network distributed consistency object state estimation method |
| CN103649738A (en) * | 2011-07-13 | 2014-03-19 | 皇家飞利浦有限公司 | Gas sensing apparatus |
| US20140375493A1 (en) * | 2012-12-28 | 2014-12-25 | Trimble Navigation Limited | Locally measured movement smoothing of gnss position fixes |
| CN105352529A (en) * | 2015-11-16 | 2016-02-24 | 南京航空航天大学 | Multisource-integrated-navigation-system distributed inertia node total-error on-line calibration method |
| CN205302636U (en) * | 2016-01-11 | 2016-06-08 | 吉林大学 | Dangerous early warning system of express delivery car |
-
2016
- 2016-12-22 CN CN201611200530.6A patent/CN106878375B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1910428A (en) * | 2003-12-05 | 2007-02-07 | 霍尼韦尔国际公司 | System and method for using multiple aiding sensors in a deeply integrated navigation system |
| CN101505532A (en) * | 2009-03-12 | 2009-08-12 | 华南理工大学 | Wireless sensor network target tracking method based on distributed processing |
| CN102221365A (en) * | 2010-04-19 | 2011-10-19 | 霍尼韦尔国际公司 | Systems and methods for determining inertial navigation system faults |
| CN103649738A (en) * | 2011-07-13 | 2014-03-19 | 皇家飞利浦有限公司 | Gas sensing apparatus |
| US20140375493A1 (en) * | 2012-12-28 | 2014-12-25 | Trimble Navigation Limited | Locally measured movement smoothing of gnss position fixes |
| CN103063212A (en) * | 2013-01-04 | 2013-04-24 | 哈尔滨工程大学 | Integrated navigation method based on non-linear mapping self-adaptive hybrid Kalman/H infinite filters |
| CN103648108A (en) * | 2013-11-29 | 2014-03-19 | 中国人民解放军海军航空工程学院 | Sensor network distributed consistency object state estimation method |
| CN105352529A (en) * | 2015-11-16 | 2016-02-24 | 南京航空航天大学 | Multisource-integrated-navigation-system distributed inertia node total-error on-line calibration method |
| CN205302636U (en) * | 2016-01-11 | 2016-06-08 | 吉林大学 | Dangerous early warning system of express delivery car |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108593557A (en) * | 2018-03-13 | 2018-09-28 | 杭州电子科技大学 | Based on TE-ANN-AWF mobile pollution source telemetry errors compensation methodes |
| CN108593557B (en) * | 2018-03-13 | 2020-08-11 | 杭州电子科技大学 | Remote measurement error compensation method based on TE-ANN-AWF (transverse electric field analysis) -based mobile pollution source |
| CN109061068A (en) * | 2018-08-15 | 2018-12-21 | 中国民航大学 | The fault-tolerant measurement estimation method of cabin pollutant concentration |
| CN109061068B (en) * | 2018-08-15 | 2019-05-21 | 中国民航大学 | The fault-tolerant measurement estimation method of cabin pollutant concentration |
| CN109459527A (en) * | 2018-09-25 | 2019-03-12 | 中国商用飞机有限责任公司 | A kind of monitoring method and monitoring system of aircraft cockpit carbonomonoxide concentration |
| CN109782269A (en) * | 2018-12-26 | 2019-05-21 | 北京壹氢科技有限公司 | A kind of distribution multi-platform cooperative active target tracking |
| CN110312225A (en) * | 2019-07-30 | 2019-10-08 | 平顶山学院 | A kind of wireless sensor hardware device |
| CN112557996A (en) * | 2019-09-26 | 2021-03-26 | 武汉国测数据技术有限公司 | Electric energy measuring system convenient for error checking and error checking method |
| CN112557996B (en) * | 2019-09-26 | 2023-11-03 | 深圳电蚂蚁数据技术有限公司 | Electric energy measurement system convenient for error verification and error verification method |
| CN111458471A (en) * | 2019-12-19 | 2020-07-28 | 中国科学院合肥物质科学研究院 | Water area detection early warning method based on graph neural network |
| CN111458471B (en) * | 2019-12-19 | 2023-04-07 | 中国科学院合肥物质科学研究院 | Water area detection early warning method based on graph neural network |
| CN111601269A (en) * | 2020-05-15 | 2020-08-28 | 中国民航大学 | Event trigger Kalman consistency filtering method based on information freshness judgment |
| CN114152810A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Three-phase electric energy sensor with three-way array structure and measuring system and method thereof |
| CN114152809B (en) * | 2020-09-08 | 2024-03-15 | 武汉国测数据技术有限公司 | Smart electric meter with error self-checking function and checking method thereof |
| CN114152811A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Electric energy meter with three-way array structure, and constituent measurement system and measurement method thereof |
| CN114152808A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Intelligent electric meter with error self-checking function and checking method thereof |
| CN114152809A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Intelligent electric meter with error self-checking function and checking method thereof |
| CN114152791A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Three-meter method three-phase electric energy meter structure for user self-checking error and checking method |
| CN114152808B (en) * | 2020-09-08 | 2024-03-15 | 武汉国测数据技术有限公司 | Smart electric meter with error self-checking function and checking method thereof |
| CN114152791B (en) * | 2020-09-08 | 2024-03-15 | 武汉国测数据技术有限公司 | Three-meter-method three-phase electric energy meter structure with user self-checking error and checking method |
| CN114152810B (en) * | 2020-09-08 | 2024-03-15 | 武汉国测数据技术有限公司 | Three-phase electric energy sensor with three-way array structure and measuring system and method thereof |
| CN114152806A (en) * | 2020-09-08 | 2022-03-08 | 武汉国测数据技术有限公司 | Electric energy sensor with three-way array structure and measurement system and method formed by electric energy sensor |
| CN114152806B (en) * | 2020-09-08 | 2024-03-15 | 武汉国测数据技术有限公司 | Electric energy sensor with three-way array structure and measurement system and method formed by same |
| CN114152811B (en) * | 2020-09-08 | 2024-03-15 | 武汉国测数据技术有限公司 | Electric energy meter with three-way array structure and measuring system and measuring method formed by electric energy meter |
| CN117241281B (en) * | 2023-11-13 | 2024-01-30 | 中赣通信(集团)有限公司 | Indoor distributed monitoring method and monitoring network |
| CN117241281A (en) * | 2023-11-13 | 2023-12-15 | 中赣通信(集团)有限公司 | Indoor distributed monitoring method and monitoring network |
| CN117713750A (en) * | 2023-12-14 | 2024-03-15 | 河海大学 | Consistency Kalman filtering state estimation method based on fractional power |
| CN117713750B (en) * | 2023-12-14 | 2024-05-17 | 河海大学 | Consistency Kalman filtering state estimation method based on fractional power |
| CN117895920A (en) * | 2024-03-13 | 2024-04-16 | 南京工业大学 | Distributed consistency Kalman filtering method for sensor network under communication link fault |
| CN117895920B (en) * | 2024-03-13 | 2024-05-17 | 南京工业大学 | Distributed consistency Kalman filtering method for sensor network under communication link fault |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106878375B (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106878375B (en) | A kind of cockpit pollutant monitoring method based on distributed combination sensor network | |
| CN106650825B (en) | Motor vehicle exhaust emission data fusion system | |
| CN109298136B (en) | Air Quality Evaluation method, apparatus, equipment and storage medium | |
| CN105243435B (en) | A kind of soil moisture content prediction technique based on deep learning cellular Automation Model | |
| CN107677997A (en) | Extension method for tracking target based on GLMB filtering and Gibbs samplings | |
| CN103017771B (en) | Multi-target joint distribution and tracking method of static sensor platform | |
| CN109615860A (en) | A kind of signalized intersections method for estimating state based on nonparametric Bayes frame | |
| CN113837361A (en) | Air pollutant concentration prediction method and system | |
| CN109254532A (en) | A kind of multiple agent cooperation fault detection method towards communication delay | |
| CN101430309B (en) | Environmental quality evaluation method based on rough set-RBF neural network | |
| CN105512011B (en) | A kind of electronics testability modeling appraisal procedure | |
| CN113591215A (en) | Abnormal satellite component layout detection method based on uncertainty | |
| CN114707632A (en) | Sensor network sensor fault positioning method, system, equipment and medium | |
| CN114298270A (en) | Pollutant concentration prediction method fusing domain knowledge and related equipment thereof | |
| CN110298409A (en) | Multi-source data fusion method towards electric power wearable device | |
| CN114048546A (en) | Graph convolution network and unsupervised domain self-adaptive prediction method for residual service life of aircraft engine | |
| Harini et al. | Statistical test for multivariate geographically weighted regression model using the method of maximum likelihood ratio test | |
| Rossini et al. | WSNs self-calibration approach for smart city applications leveraging incremental machine learning techniques | |
| CN103796217B (en) | A kind of estimation range partitioning method and device based on drive test data | |
| CN113344759B (en) | Analysis method for pollution emission of mobile source | |
| CN104467742A (en) | Sensor network distribution type consistency particle filter based on Gaussian mixture model | |
| CN115526410A (en) | Method for predicting atmospheric pollutant data based on multi-parameter spatial filtering prediction model | |
| CN105373673B (en) | A kind of natural electric field monitoring data dynamic playback method and system | |
| Georges | Energy minimization and observability maximization in multi-hop wireless sensor networks | |
| Mediero et al. | Improving probabilistic flood forecasting through a data assimilation scheme based on genetic programming |
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 |