CN102005105A - Marine disaster early-warning device based on time series similarity matching - Google Patents
Marine disaster early-warning device based on time series similarity matching Download PDFInfo
- Publication number
- CN102005105A CN102005105A CN201010292792.6A CN201010292792A CN102005105A CN 102005105 A CN102005105 A CN 102005105A CN 201010292792 A CN201010292792 A CN 201010292792A CN 102005105 A CN102005105 A CN 102005105A
- Authority
- CN
- China
- Prior art keywords
- oceanic disasters
- sequence
- disasters
- oceanic
- time
- 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
Images
Landscapes
- Position Fixing By Use Of Radio Waves (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a marine disaster early-warning device based on time series similarity matching, a global positioning system (GPS)/ a gyroscope and an electronic compass. The marine disaster early-warning device comprises a GPS module, the gyroscope, the electronic compass, video acquisition equipment, multi-touch equipment, a data processing device, a sensor, a geographic information system (GIS) and a location-based service (LBS), an alarm device and the like. The GPS module, the gyroscope and the electronic compass form a locating device to acquire the locating information. The GIS and LBS location-based service is used for precisely determining the geographic location and overlay much real world information. The data processing device is used for storing characteristic data recorded during the occurrence of the history marine disaster and a processing method, and informs that whether the marine disaster early-warning device is in the danger zone according to the real-time data acquired by the sensor by a similarity matching method. Finally, the multi-touch hand-held equipment sends alarming sound when the marine disaster early-warning device is in the danger zone and the screen displays the red zone. The user can find the safe zone by using multi-point sliding and retreat and guide to retreat. The marine disaster early-warning device is used for marine operators, common fishermen and researchers, serves as an aid decision-making manner for leaders, has real-time property and is easy to carry.
Description
Technical field
The present invention relates to the guider of a kind of GPS of utilization, gyroscope, digital compass combination, and use sensor to receive the early warning of marine disasters device that data are carried out the similar coupling of time series.
Background technology
The ocean is the resource treasure-house, be the food resource base, be the important energy base, be the mineral metal products resources base, be the water resource base, be renewable sources of energy base, be productive life space resources, be the thoroughfare of world's traffic.And nearest variation owing to expanding economy and environment, economic loss and casualties that etesian Oceanic disasters cause to the coastland, economic construction and the construction of harmonious society to the coastland constitutes a serious threat.Therefore, the research early warning technology is strengthened coastland Oceanic disasters monitoring, obtains the ocean instant messages, and the development decision support system (DSS) of taking precautions against natural calamities is issued the condition of a disaster accurately, is to assure the safety for life and property of the people and the pressing for of social stability.
Time Series Similarity coupling has great importance at aspects such as statistical theory, machine learning and data minings, and similarity measurement is usually used in finding to have the stock of similar fluctuating, the product of determining to have similar sales mode, classification and has stellar spectrum curve of similar shape or the like.The historical data of Oceanic disasters is abundanter, and making full use of these data is powerful measure that ensure people's life safety.Locating device uses the most extensive in auto navigation, and technology is comparative maturity also, and combines based on reality technology at present, and the real world content better superposes.
Summary of the invention
The purpose of this invention is to provide a kind of early warning of marine disasters hand-held device, be particularly suited for the prior-warning device of storm surge disaster, make things convenient for the user to carry based on the similarity coupling.Be mainly used in offshore operation personnel and researcher.Under the background of the fast development of Global climate change and coastal economy, Oceanic disasters occurrence frequency and intensity highlight, as strange tide of storm tide, Nantong etc.
In order to realize above purpose, the present invention adopts following technical scheme to realize:
1) locating device: receive information by digital compass, MEMS gyroscope and GPS, when gps signal is blocked, receive information by gyroscope and digital compass, simultaneously, the MEMS gyro provides the attitude of handheld device, reads different scene content according to the different directions of equipment; The accurate direction of the current position of digital compass indication user; The accurate direction of GPS consumer positioning present position.Obtain the longitude and latitude of object of correction handheld device by the LBS location-based service, thereby determine user's particular location.
2) acquisition user's scene content: the user absorbs the scene that locating device and LBS located by video capture device, and is presented on the multiple point touching handheld device.
3) use of sensor: relevant on the Oceanic disasters more than temperature, air pressure etc., so the present invention determines to have temperature, two sensors of air pressure at least, can increase the kind of sensor as required again.Sensor receives the data of returning and passes to data processing equipment, compares with case over the years, provides a threshold value, if similarity more than or equal to this threshold value, is then thought the hazardous location.Pass to signal of warning device, inform that this place is the hazardous location, and show red area on the handheld device.
4) data processing equipment: data processing equipment is very crucial in the present invention, and its receives the data from sensor, characteristic and disposal route when having disaster over the years simultaneously and taking place.Step is as follows:
A, with the Oceanic disasters catalogue data be converted in certain space-time unique, Oceanic disasters event train collection more than certain rank.Like this, the interior Oceanic disasters of different space-time uniques have just constituted different Oceanic disasters sequences.Provide the definition of Oceanic disasters sequence and similarity thereof and the measurement model of Oceanic disasters sequence similarity below.
Definition 1 (the relevant area of Oceanic disasters): frequent geographic area takes place in Oceanic disasters.
Definition 2 (Oceanic disasters incidents). regard the Oceanic disasters each time that write down in the Oceanic disasters catalog data base as an incident.All Oceanic disasters incidents are divided into different Oceanic disasters event set by space attribute with the zone, the Oceanic disasters incident that occurs in more than certain rank and the while takes place constantly at t is designated as E (t).
Definition 3 (Oceanic disasters sequences of events). in the relevant area of certain Oceanic disasters, the event set Et={E (t) that occurs in the time range F | the arrangement of t ∈ time range F} on time shaft calls the Oceanic disasters sequence of events, and wherein each independent incident E (t) is called an event items.
Definition 4 (Oceanic disasters sequences). according to time and the grade that Oceanic disasters take place, the space attribute of Oceanic disasters event items is omitted, abstract on the time shaft be a storm tide intensity level in the chronomere.Thereby an Oceanic disasters sequence of events just is conceptualized as a time series, is called the Oceanic disasters sequence, with S={S (t) | and t=0,1,2 ..., n} represents, in the Oceanic disasters sequence of events, and an element in corresponding Oceanic disasters sequence of event items.What will specify here is that according to the field characteristic, the time t that the Oceanic disasters incident is taken place expands to a unit interval scope, i.e. t ∈ Tgap.
Definition 5 (Oceanic disasters sequence lengths). for Oceanic disasters sequence S, use | S| represents its length, i.e. the element number of Oceanic disasters sequence S.S[i] represent i element among the Oceanic disasters sequence S, S is in t grade value constantly for S (t) expression Oceanic disasters sequence.
Definition 6 (dense Oceanic disasters sequences). in zone sometime,, replenished the Oceanic disasters sequence that empty element forms later on, be called dense Oceanic disasters sequence S ' in the unit interval section that the Oceanic disasters incident does not take place to Oceanic disasters sequence S.Annotate: following said sequence and element all are meant the element among dense sequence or the dense sequence.
Definition 7 (Oceanic disasters similaritys). establish x and y and be the corresponding element in any two different Oceanic disasters sequences, (x y) describes the similarity of x and y element, and the distance between element x, the y is used dynamic bending distance (DTW) with labeling function A, be designated as: D (x, y).Labeling function A (x y) is defined as:
If D (x, y)≤M
Threshold, then A (x, y)=0; If D (x, y)＞M
Threshold, then A (x, y)=1.Here, M
ThresholdBe error threshold, determine according to real needs by the user.Error threshold is more little, and the Oceanic disasters intensity difference of two element representatives of expression is more little.It should be noted that especially that because allow the time that relevant Oceanic disasters take place that certain interval is arranged so x and y two elements can corresponding different time points, the user is the scope in given this time interval as required.
On the other hand, usually such phenomenon can appear in each disaster sequence of ocean: in some unit interval section of time shaft, do not have the Oceanic disasters incident to take place.At this moment, can replenish empty element in sequence, be 0 element in additional value of the time period that does not have the Oceanic disasters incident promptly.When non-NULL element and empty element carry out similarity relatively the time, the labeling function value is 0.
Definition 8 (Oceanic disasters similarity measurement models). establishing S is two different Oceanic disasters sequences with Q, and then the similarity of S and R can and be measured with the similarity labeling function in two sequences.It is defined as follows:
Defined for 9 (Oceanic disasters sequence piecewise linearity is represented). in the certain hour scope, the Oceanic disasters sequence is expressed as the adjacent near linear of multistage according to time t.Because the data of Oceanic disasters have the feature that the part peels off, this linear segmented adopts the method for moving window.Be that Oceanic disasters sequence of events s and the length that a given length is n is the time window of w, note s is (x1, x2, ..., xn), time window is placed on the seasonal effect in time series reference position, length is the cross-talk sequence of w on the corresponding sequence of time window this moment, and the like, be divided into section into n-w+1.
Definition 10 (relevant Oceanic disasters number of support and supports). relevant Oceanic disasters number of support Sup (S, Q) and support η be defined as
η＝Sup(S，Q)/|S|
Wherein, E (t) ∈ { S
t, Q
tBe illustrated in the time interval scope t ∈ [T-Tgap, T+Tgap], as long as all there is the Oceanic disasters incident to take place among sequence S and the sequence Q, the support sum just increases by 1; There is one not have the Oceanic disasters incident as infructescence S and sequence Q, just do not charge to the support sum.N in the formula (2) is a time interval sum.
The sequence similarity matching algorithm in B, the relevant area of searching Oceanic disasters
Characteristics in conjunction with relevant Oceanic disasters, defining that Oceanic disasters sequence piecewise linearity is represented, on the basis of the support of relevant Oceanic disasters and Oceanic disasters sequence similarity measurement model, proposing a kind of whole sequence similarity matching algorithm based on number of support and linear segmented.Provide the step of algorithm below:
Step 1: pre-service---from the Oceanic disasters catalogue data, take out data,, be divided into different zones, and all together leave in the corresponding informance in longitude and latitude and zone in the pre-service destination file by the locus according to its latitude and longitude information that comprises;
Step 2: constitute the Oceanic disasters sequence---from the pre-service destination file, the aggregation of data that extraction has the same area numbering arrives together, put into different temporary files respectively, and the Oceanic disasters record strip number in this zone of accumulative total, charge in (regional number, bar number) list structure.At last, the case history in each zone according to time sequence just forms one group of Oceanic disasters time series EQS;
Step 3: the sequence piecewise linearity represents---the method that Oceanic disasters time series EQS is divided according to moving window is divided into the n-w+1 section with sequence, as defines shown in 9.
Step 4: carry out in twos that similarity analysis---detailed process is: take out the Oceanic disasters time series that is in the zones of different respectively,, then make corresponding record if the similarity of two sequences reaches requirement.Calculate support then, and whether reach threshold value output correlated series and check in this relevant history and handle record according to support;
Step 5: amalgamation result---the relevant area of Oceanic disasters is merged according to the geographic position, and recomputate support.
When in this algorithm, two sequences being carried out similarity determination, need one by one element relatively.
A given as required threshold value, the data of utilizing sensor to receive are calculated, if the gained result is more than or equal to this threshold value, then judge it is the hazardous location, signal is passed to warning device, simultaneously, if the user is in the hazardous location, then on touch-screen, consult a map, select to withdraw from nearest place, this hazardous location.If managerial personnel use, can also further determine the most approaching of itself and historical data which year, the method for dangerous situation is handled in reference then, harm is dropped to minimum.
5) warning device comprises a colour switching device and a sound-producing device, and according to the result of data processing equipment, warning device gives different responses, and when having arrived the hazardous location, sonification system gives the alarm, and this zone becomes redness simultaneously.
6) sensor device in time receives data such as temperature, air pressure, water temperature and passes to data processing equipment and handle.
Description of drawings
Fig. 1 is the structural drawing of ocean of the present invention prior-warning device.
Fig. 2 is the logic function figure of invention ocean prior-warning device.
Embodiment
The invention discloses a kind of method and analysis of the early warning of marine disasters device based on the similar coupling of time series, embodiment is described below in conjunction with accompanying drawing.
Figure 1 shows that the building-block of logic of early warning of marine disasters device of the present invention, comprise that locating device 102 is (by digital compass, MEMS gyroscope and GPS form), LBS and GPS technology are in conjunction with 105, data processing equipment 104: deposit characteristic and the disposal route put down in writing when Oceanic disasters over the years take place, warning device 106: the data of sending back according to data processing equipment judge whether to produce alerting signal, sensor device 107: the data of receive are passed to data processing equipment handle, video capture device 103: obtain the scene of locating device picked-up and be presented on the multi-point touch handheld device and multi-point touch handheld device 101 by this equipment.Different with traditional early warning of marine disasters device, this device has real-time, and easy to carry, uses object more extensive.By service of LBS position-based and video capture device, on real world, the stack more information has the advantages that reality strengthens.Wherein:
Locating device 102 is used to receive position and direction and the current being seen scope that positioning signal is determined the user place.The GPS module is used for receiving satellite positioning signals, real-time positioning, and positioning result is kept in the locating device; The digital compass module is used to indicate the residing direction of prior-warning device, and shows different directions along with the transformation of gyroscope attitude; Gyro module, assistant GPS provide orientation accurately, when the GPS received signal is unstable, can be used as locating device mobile message is provided.
Data processing equipment 104 is used to receive the data from sensor, characteristic and disposal route when having disaster over the years simultaneously and taking place.Utilizing the method for Time Series Similarity coupling, is example (can be any one of Oceanic disasters, as typhoon, strange tide etc.) with the ocean storm tide below, and step is as follows:
A, with the storm tide catalogue data be converted in certain space-time unique, storm tide event train collection more than certain rank. like this, the storm tide in the different space-time uniques has just constituted different storm tide sequences. the following definition of storm tide sequence and similarity thereof and the measurement model of storm tide sequence similarity of providing.
Definition 1 (the relevant area of storm tide): frequent geographic area takes place in storm tide.
Definition 2 (storm tide incidents). regard the storm tide each time that writes down in the storm tide catalog data base as an incident.All storm tide incidents are divided into different storm tide event set by space attribute with the zone, the storm tide incident that occurs in more than certain rank and the while takes place constantly at t is designated as E (t).
Definition 3 (storm tide sequences of events). in the relevant area of certain storm tide, the event set E that occurs in the time range F
t=E (t) | the arrangement of t ∈ time range F} on time shaft is called the storm tide sequence of events, and wherein each independent incident E (t) is called an event items.
Definition 4 (storm tide sequences). according to time and the grade that storm tide takes place, the space attribute of storm tide event items is omitted, abstract on the time shaft be a storm tide intensity level in the chronomere.Thereby a storm tide sequence of events just is conceptualized as a time series, is called the storm tide sequence, with S={S (t)=0,1,2 ..., n} represents, in the storm tide sequence of events, and an element in corresponding storm tide sequence of event items.What will specify here is that according to the field characteristic, the time t that the storm tide incident is taken place expands to a unit interval scope, i.e. t ∈ T
Gap
Definition 5 (storm tide sequence lengths). for storm tide sequence S, use | S| represents its length, i.e. the element number of storm tide sequence S.S
[i]I element among the expression storm tide sequence S, S is in t grade value constantly for S (t) expression storm tide sequence.
Definition 6 (dense storm tide sequences). in zone sometime,, replenished the storm tide sequence that empty element forms later on, be called dense storm tide sequence S ' in the unit interval section that the storm tide incident does not take place to storm tide sequence S.Annotate: following said sequence and element all are meant the element among dense sequence or the dense sequence.
Definition 7 (storm tide similaritys). establish x and y and be the corresponding element in any two different storm tide sequences, (x y) describes the similarity of x and y element, and the distance between element x, the y is used dynamic bending distance (DTW) with labeling function A, be designated as: D (x, y).Labeling function A (x y) is defined as:
If D (x, y)≤M
Threshold, then A (x, y)=0; If D (x, y)＞M
Threshold, then A (x, y)=1.Here, M
ThresholdBe error threshold, determine according to real needs by the user.Error threshold is more little, and the storm tide intensity difference of two element representatives of expression is more little.It should be noted that especially that because allow the time that relevant storm tide takes place that certain interval is arranged so x and y two elements can corresponding different time points, the user is the scope in given this time interval as required.
On the other hand, usually such phenomenon can appear in each disaster sequence of ocean: in some unit interval section of time shaft, do not have the Oceanic disasters incident to take place.At this moment, can replenish empty element in sequence, be 0 element in additional value of the time period that does not have the Oceanic disasters incident promptly.When non-NULL element and empty element carry out similarity relatively the time, the labeling function value is 0.
Definition 8 (storm tide similarity measurement models). establishing S is two different storm tide sequences with Q, and then the similarity of S and R can and be measured with the similarity labeling function in two sequences.It is defined as follows:
Defined for 9 (storm tide sequence piecewise linearity is represented). in the certain hour scope, the storm tide sequence is expressed as the adjacent near linear of multistage according to time t.Because the data of Oceanic disasters have the feature that the part peels off, this linear segmented adopts the method for moving window.Be that storm tide sequence of events s and the length that a given length is n is the time window of w, note s is (x1, x2, ..., xn), time window is placed on the seasonal effect in time series reference position, length is the cross-talk sequence of w on the corresponding sequence of time window this moment, and the like, be divided into section into n-w+1.
Definition 10 (relevant storm tide number of support and supports). relevant storm tide number of support Sup (S, Q) and support η be defined as
η＝Sup(S，Q)/|S|
Wherein, E (t) ∈ { S
t, Q
tBe illustrated in the time interval scope t ∈ [T-Tgap, T+Tgap], as long as all there is the storm tide incident to take place among sequence S and the sequence Q, the support sum just increases by 1; There is one not have the storm tide incident as infructescence S and sequence Q, just do not charge to the support sum.N in the formula (2) is a time interval sum.
The sequence similarity matching algorithm in B, the relevant area of searching storm tide
The sequence similarity coupling can be divided into two classes: a class is a whole matching, promptly given m the data sequence that length is n, and a search sequence and the tolerance ε that length is n finds out the data sequence similar to search sequence from m data sequence; Another kind of is the subsequence coupling, search sequence is shorter than the records series in the database, need in records series, seek the subsequence similar with search sequence, also be given m data sequence, a search sequence and a tolerance ε, from m data sequence, find out its subsequence data sequence similar, and provide the side-play amount of subsequence in data sequence to search sequence.Search the relevant area of storm tide with the sequence similarity Matching Algorithm, main thought is exactly in each regional storm tide sequence, seeks the storm tide sequence with high similarity with the method for whole sequence similarity coupling.For this reason, we are in conjunction with the characteristics of relevant storm tide, defining that storm tide sequence piecewise linearity is represented, on the basis of the support of relevant storm tide and storm tide sequence similarity measurement model, proposing a kind of whole sequence similarity matching algorithm based on number of support and linear segmented.Provide the step of algorithm below:
Step (1): pre-service---from the storm tide catalogue data, take out data,, be divided into different zones, and all together leave in the corresponding informance in longitude and latitude and zone in the pre-service destination file by the locus according to its latitude and longitude information that comprises;
Step (2): constitute the storm tide sequence---from the pre-service destination file, the aggregation of data that extraction has the same area numbering arrives together, put into different temporary files respectively, and the storm tide record strip number in this zone of accumulative total, charge in (regional number, bar number) list structure.At last, the case history in each zone according to time sequence just forms one group of storm tide time series EQS;
Step (3): the sequence piecewise linearity represents---the method that storm tide time series EQS is divided according to moving window is divided into the n-w+1 section with sequence, as defines shown in 9.
Step (4): carry out in twos that similarity analysis---detailed process is: take out the storm tide time series that is in the zones of different respectively,, then make corresponding record if the similarity of two sequences reaches requirement.Calculate support then, and whether reach threshold value output correlated series and check in this relevant history and handle record according to support;
Step (5): amalgamation result---the relevant area of storm tide is merged according to the geographic position, and recomputate support.
When in this algorithm, two sequences being carried out similarity determination, need one by one element relatively.
A given as required threshold value, the data of utilizing sensor to receive are calculated, if the gained result is more than or equal to this threshold value, then judge it is the hazardous location, signal is passed to warning device, simultaneously, if the user is in the hazardous location, then on touch-screen, consult a map, select to withdraw from nearest place, this hazardous location.If managerial personnel use, can also further determine the most approaching of itself and historical data which year, the method for dangerous situation is handled in reference then, harm is dropped to minimum.
Warning device 106 comprises a colour switching device and a sound-producing device, and according to the result of data processing equipment, warning device gives different responses, and when having arrived the hazardous location, sonification system gives the alarm, and this zone becomes redness simultaneously.
Sensor device 107 in time receives data such as temperature, air pressure, water temperature and passes to data processing equipment and handle.
Video capture device 103 is used to obtain the scope that locating device absorbs, thereby superposes more information on real world.
Multi-point touch handheld device 101, use GIS and location information service to combine, some required GIS technology (as information such as the spheroid among the GIS, maps) is transplanted on the handheld device, utilize space orientation technique and GIS technology, the positional information obtained and other spaces and attribute information are collected, select required information for the user.
Fig. 2 is the logic function figure of early warning of marine disasters device of the present invention, locating device utilizes GPS or gyroscope to obtain positional information, digital compass indicates the residing direction of device, gyroscope is provided with different attitudes according to the change of design factors, and promptly under different attitudes, the range content that is absorbed is different, as in same bay district, there are whirlpool or gully in possible front, and the geographical location information that GPS shows is identical, but gyroscope can be so that the scope of picked-up be more accurate.GIS and the combination of LBS technology, some required GIS technology (as information such as the spheroid among the GIS, maps) is transplanted on the handheld device, utilize space orientation technique and GIS technology, the positional information obtained and other spaces and attribute information are collected, select required information for the user.Video capture device is filmed the being seen scene content of location-based service of current locating device and GIS, LBS combination and be presented on the multiple point touching handheld device, thereby superposes more information on real world, has the service of augmented reality.If the user is in the dangerous landform of the Oceanic disasters zone, then carry out early warning, show this zone on the handheld device immediately for red, and sound the alarm.If the user is not in the dangerous landform of Oceanic disasters zone, then according to the data of sensor, utilize data processing equipment to calculate, judge whether this zone disaster can take place, as the ocean storm tide, can calculate according to data such as temperature, air pressure, the over the years characteristic similar coupling of real time data to data processing device for storage, if the end value of coupling is more than or equal to the threshold value that provides, then carry out early warning, show this zone on the multi-point touch handheld device immediately for red, and sound the alarm.
In sum, the present invention is a kind of early warning of marine disasters device based on the similar coupling of time series, data processing equipment adopts the real time data that sensor obtains the method for similar coupling to judge whether the user is in the danger zone, locating device provides direction accurately, and is presented on the multi-point touch handheld device.
The above only is an illustrative, but not is restricted, and is unrestricted as the MEMS gyroscope, can be other type, and number of sensors and kind also can be determined by the user etc. voluntarily.This description should not be construed as limitation of the present invention, and any spirit of the present invention and category of not breaking away from all should be contained within the application range.
Claims (8)
1. the method based on the early warning of marine disasters device of the similar coupling of time series is characterized in that, should comprise: a location device 102, form by a digital compass, a MEMS gyroscope and a GPS;
One data processing equipment 104, the characteristic and the disposal route of record when depositing Oceanic disasters over the years and taking place, and the method for mating with the time sequence similarity;
One warning device 106, the data of sending back according to data processing equipment judge whether to produce alerting signal;
One sensor device 107 is passed to data processing equipment to the data of receive and is handled;
One LBS, GPS binding modules 105 are embedded into the part of GPS technology on the multi-point touch handheld device, in conjunction with the LBS location-based service, obtain the required details of more users;
One video capture device 103 obtains the scene of locating device picked-up and be presented on the multi-point touch handheld device by this equipment.
One multi-point touch handheld device 101, this equipment has the function of multiple point touching, utilizes locating device and video capture device and LBS, GPS combination technology, just can judge the ad-hoc location and the canned data at user place.
2. locating device according to claim 1 is characterized in that, during the GPS positioning stablity, and the accurate direction of GPS consumer positioning present position, the MEMS gyro provides the attitude of handheld device, reads different scene content according to the different directions of equipment; The accurate direction of the current position of digital compass indication user; When GPS can not accurately locate, determine by MEMS gyroscope and digital compass.
3. data processing equipment according to claim 1 is characterized in that, utilizes the method for Time Series Similarity coupling, and step is as follows:
A, with the Oceanic disasters catalogue data be converted in certain space-time unique, Oceanic disasters event train collection more than certain rank.Like this, the interior Oceanic disasters of different space-time uniques have just constituted different Oceanic disasters sequences.Provide the definition of Oceanic disasters sequence and similarity thereof and the measurement model of Oceanic disasters sequence similarity below.
Definition 1 (the relevant area of Oceanic disasters): frequent geographic area takes place in Oceanic disasters.
Definition 2 (Oceanic disasters incidents). regard the Oceanic disasters each time that write down in the Oceanic disasters catalog data base as an incident.All Oceanic disasters incidents are divided into different Oceanic disasters event set by space attribute with the zone, the Oceanic disasters incident that occurs in more than certain rank and the while takes place constantly at t is designated as E (t).
Definition 3 (Oceanic disasters sequences of events). in the relevant area of certain Oceanic disasters, the event set Et={E (t) that occurs in the time range F | the arrangement of t ∈ time range F} on time shaft calls the Oceanic disasters sequence of events, and wherein each independent incident E (t) is called an event items.
Definition 4 (Oceanic disasters sequences). according to time and the grade that Oceanic disasters take place, the space attribute of Oceanic disasters event items is omitted, abstract on the time shaft be a storm tide intensity level in the chronomere.Thereby an Oceanic disasters sequence of events just is conceptualized as a time series, is called the Oceanic disasters sequence, with S={S (t) | and t=0,1,2 ..., n} represents, in the Oceanic disasters sequence of events, and an element in corresponding Oceanic disasters sequence of event items.What will specify here is that according to the field characteristic, the time t that the Oceanic disasters incident is taken place expands to a unit interval scope, i.e. t ∈ Tgap.
Definition 5 (Oceanic disasters sequence lengths). for Oceanic disasters sequence S, use | S| represents its length, i.e. the element number of Oceanic disasters sequence S.S[i] represent i element among the Oceanic disasters sequence S, S is in t grade value constantly for S (t) expression Oceanic disasters sequence.
Definition 6 (dense Oceanic disasters sequences). in zone sometime,, replenished the Oceanic disasters sequence that empty element forms later on, be called dense Oceanic disasters sequence S ' in the unit interval section that the Oceanic disasters incident does not take place to Oceanic disasters sequence S.Annotate: following said sequence and element all are meant the element among dense sequence or the dense sequence.
Definition 7 (Oceanic disasters similaritys). establish x and y and be the corresponding element in any two different Oceanic disasters sequences, (x y) describes the similarity of x and y element, and the distance between element x, the y is used dynamic bending distance (DTW) with labeling function A, be designated as: D (x, y).Labeling function A (x y) is defined as:
If D (x, y)≤M
Threshold, then A (x, y)=0; If D (x, y)＞M
Threshold, then A (x, y)=1.Here, M
ThresholdBe error threshold, determine according to real needs by the user.Error threshold is more little, and the Oceanic disasters intensity difference of two element representatives of expression is more little.It should be noted that especially that because allow the time that relevant Oceanic disasters take place that certain interval is arranged so x and y two elements can corresponding different time points, the user is the scope in given this time interval as required.
On the other hand, usually such phenomenon can appear in each disaster sequence of ocean: in some unit interval section of time shaft, do not have the Oceanic disasters incident to take place.At this moment, can replenish empty element in sequence, be 0 element in additional value of the time period that does not have the Oceanic disasters incident promptly.When non-NULL element and empty element carry out similarity relatively the time, the labeling function value is 0.
Definition 8 (Oceanic disasters similarity measurement models). establishing S is two different Oceanic disasters sequences with Q, and then the similarity of S and R can and be measured with the similarity labeling function in two sequences.It is defined as follows:
Defined for 9 (Oceanic disasters sequence piecewise linearity is represented). in the certain hour scope, the Oceanic disasters sequence is expressed as the adjacent near linear of multistage according to time t.Because the data of Oceanic disasters have the feature that the part peels off, this linear segmented adopts the method for moving window.Be that Oceanic disasters sequence of events s and the length that a given length is n is the time window of w, note s is (x1, x2, ..., xn), time window is placed on the seasonal effect in time series reference position, length is the cross-talk sequence of w on the corresponding sequence of time window this moment, and the like, be divided into section into n-w+1.
Definition 10 (relevant Oceanic disasters number of support and supports). relevant Oceanic disasters number of support Sup (S, Q) and support η be defined as
η＝Sup(S，Q)/|S|
Wherein, E (t) ∈ { S
t, Q
tBe illustrated in the time interval scope t ∈ [T-Tgap, T+Tgap], as long as all there is the Oceanic disasters incident to take place among sequence S and the sequence Q, the support sum just increases by 1; There is one not have the Oceanic disasters incident as infructescence S and sequence Q, just do not charge to the support sum.N in the formula (2) is a time interval sum.
The sequence similarity matching algorithm in B, the relevant area of searching Oceanic disasters
Characteristics in conjunction with relevant Oceanic disasters, defining that Oceanic disasters sequence piecewise linearity is represented, on the basis of the support of relevant Oceanic disasters and Oceanic disasters sequence similarity measurement model, proposing a kind of whole sequence similarity matching algorithm based on number of support and linear segmented.Provide the step of algorithm below:
Step (1): pre-service---from the Oceanic disasters catalogue data, take out data,, be divided into different zones, and all together leave in the corresponding informance in longitude and latitude and zone in the pre-service destination file by the locus according to its latitude and longitude information that comprises;
Step (2): constitute the Oceanic disasters sequence---from the pre-service destination file, the aggregation of data that extraction has the same area numbering arrives together, put into different temporary files respectively, and the Oceanic disasters record strip number in this zone of accumulative total, charge in (regional number, bar number) list structure.At last, the case history in each zone according to time sequence just forms one group of Oceanic disasters time series EQS;
Step (3): the sequence piecewise linearity represents---the method that Oceanic disasters time series EQS is divided according to moving window is divided into the n-w+1 section with sequence, as defines shown in 9.
Step (4): carry out in twos that similarity analysis---detailed process is: take out the Oceanic disasters time series that is in the zones of different respectively,, then make corresponding record if the similarity of two sequences reaches requirement.Calculate support then, and whether reach threshold value output correlated series and check in this relevant history and handle record according to support;
Step (5): amalgamation result---the relevant area of Oceanic disasters is merged according to the geographic position, and recomputate support.
When in this algorithm, two sequences being carried out similarity determination, need one by one element relatively.
4. sensor device according to claim 1 is characterized in that comprising at least temperature sensor and baroceptor.
5. video capture device according to claim 1 is characterized in that the scope that current locating device is absorbed gets access on the handheld device.
6. multi-point touch handheld device according to claim 1 is characterized in that USB interface is arranged, sound-producing device, and the screen handoff functionality supports convergent-divergent little, in light weight with hand-held and volume, suitable portable carrying.
7. according to claim 1, it is characterized in that the multi-point touch handheld device utilizes GIS and location information service to combine, promptly some required GIS technology transplant on handheld device, utilize space orientation technique and GIS technology, the positional information obtained and other spaces and attribute information are collected, select required information for the user.
8. according to claim 2, it is characterized in that adopting MEMS gyroscope and electronic guide to replenish and make it more accurate at GPS location, when gps signal is blocked, can use the gyrostatic mobile message of MEMS to carry out the guiding in path, being aided with digital compass again points the direction, thereby the navigation of whole prior-warning device is very stable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102927926A CN102005105B (en) | 2010-09-27 | 2010-09-27 | Marine disaster early-warning device based on time series similarity matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102927926A CN102005105B (en) | 2010-09-27 | 2010-09-27 | Marine disaster early-warning device based on time series similarity matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102005105A true CN102005105A (en) | 2011-04-06 |
| CN102005105B CN102005105B (en) | 2012-10-31 |
Family
ID=43812433
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010102927926A Expired - Fee Related CN102005105B (en) | 2010-09-27 | 2010-09-27 | Marine disaster early-warning device based on time series similarity matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102005105B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077582A (en) * | 2011-10-25 | 2013-05-01 | 成都玺汇科技有限公司 | Novel monitoring and early warning system for sea storm |
| CN103294911A (en) * | 2013-05-23 | 2013-09-11 | 中国人民解放军国防科学技术大学 | Time sequence similarity value acquisition method and system |
| TWI420426B (en) * | 2011-05-05 | 2013-12-21 | Chunghwa Telecom Co Ltd | Multidimensional emergency messaging system |
| CN103914461A (en) * | 2012-12-31 | 2014-07-09 | 北京中交兴路信息科技有限公司 | Traffic information query method and device |
| CN106874674A (en) * | 2017-02-20 | 2017-06-20 | 上海海洋大学 | A kind of multivariate time series method for measuring similarity towards marine field |
| CN107784311A (en) * | 2016-08-24 | 2018-03-09 | 中国海洋大学 | Global mesoscale eddy space-time hierarchical topology path construction technology |
| CN109196563A (en) * | 2016-06-08 | 2019-01-11 | Sts国防有限公司 | Predicted temperature rise event |
| CN110662165A (en) * | 2019-10-09 | 2020-01-07 | 广州爱浦路网络技术有限公司 | Method for selecting user plane function UPF based on node number |
| CN110766054A (en) * | 2019-09-23 | 2020-02-07 | 东软集团股份有限公司 | Early warning method and device for fracturing sand plugging, storage medium and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6549165B2 (en) * | 2001-01-19 | 2003-04-15 | Agence Spatiale Europeenne | Ocean altimetry interferometric method and device using GNSS signals |
| US20070296575A1 (en) * | 2006-04-29 | 2007-12-27 | Trex Enterprises Corp. | Disaster alert device, system and method |
| CN101169805A (en) * | 2007-12-11 | 2008-04-30 | 肖盛燮 | Reservoir geological calamity evolution process dynamic tracking method |
| CN201084256Y (en) * | 2007-08-10 | 2008-07-09 | 河北省气象科技服务中心 | A calamity warning system |
| CN101358867A (en) * | 2008-09-03 | 2009-02-04 | 山东省科学院海洋仪器仪表研究所 | Ocean water level real-time monitoring system |
-
2010
- 2010-09-27 CN CN2010102927926A patent/CN102005105B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6549165B2 (en) * | 2001-01-19 | 2003-04-15 | Agence Spatiale Europeenne | Ocean altimetry interferometric method and device using GNSS signals |
| US20070296575A1 (en) * | 2006-04-29 | 2007-12-27 | Trex Enterprises Corp. | Disaster alert device, system and method |
| CN201084256Y (en) * | 2007-08-10 | 2008-07-09 | 河北省气象科技服务中心 | A calamity warning system |
| CN101169805A (en) * | 2007-12-11 | 2008-04-30 | 肖盛燮 | Reservoir geological calamity evolution process dynamic tracking method |
| CN101358867A (en) * | 2008-09-03 | 2009-02-04 | 山东省科学院海洋仪器仪表研究所 | Ocean water level real-time monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 《华东师范大学学报(自然科学版)》 19971231 孔亚珍 《海洋时间序列的谱分析方法》 , 2 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI420426B (en) * | 2011-05-05 | 2013-12-21 | Chunghwa Telecom Co Ltd | Multidimensional emergency messaging system |
| CN103077582A (en) * | 2011-10-25 | 2013-05-01 | 成都玺汇科技有限公司 | Novel monitoring and early warning system for sea storm |
| CN103914461B (en) * | 2012-12-31 | 2017-12-12 | 北京中交兴路信息科技有限公司 | Traffic information enquiry method and device |
| CN103914461A (en) * | 2012-12-31 | 2014-07-09 | 北京中交兴路信息科技有限公司 | Traffic information query method and device |
| CN103294911B (en) * | 2013-05-23 | 2016-12-28 | 中国人民解放军国防科学技术大学 | A kind of time series similarity value-acquiring method and system |
| CN103294911A (en) * | 2013-05-23 | 2013-09-11 | 中国人民解放军国防科学技术大学 | Time sequence similarity value acquisition method and system |
| CN109196563A (en) * | 2016-06-08 | 2019-01-11 | Sts国防有限公司 | Predicted temperature rise event |
| CN109196563B (en) * | 2016-06-08 | 2021-07-20 | Sts国防有限公司 | Predicting temperature rise events |
| CN107784311A (en) * | 2016-08-24 | 2018-03-09 | 中国海洋大学 | Global mesoscale eddy space-time hierarchical topology path construction technology |
| CN106874674A (en) * | 2017-02-20 | 2017-06-20 | 上海海洋大学 | A kind of multivariate time series method for measuring similarity towards marine field |
| CN106874674B (en) * | 2017-02-20 | 2019-04-19 | 上海海洋大学 | A kind of multivariate time series method for measuring similarity towards marine field |
| CN110766054A (en) * | 2019-09-23 | 2020-02-07 | 东软集团股份有限公司 | Early warning method and device for fracturing sand plugging, storage medium and equipment |
| CN110766054B (en) * | 2019-09-23 | 2022-05-10 | 东软集团股份有限公司 | Early warning method and device for fracturing sand plugging, storage medium and equipment |
| CN110662165A (en) * | 2019-10-09 | 2020-01-07 | 广州爱浦路网络技术有限公司 | Method for selecting user plane function UPF based on node number |
| CN110662165B (en) * | 2019-10-09 | 2020-12-15 | 广州爱浦路网络技术有限公司 | Method for selecting user plane function UPF based on node number |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102005105B (en) | 2012-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102005105B (en) | Marine disaster early-warning device based on time series similarity matching | |
| US20230168099A1 (en) | Mobile search based on predicted location | |
| Gong et al. | Deriving personal trip data from GPS data: A literature review on the existing methodologies | |
| Feng et al. | A survey on trajectory data mining: Techniques and applications | |
| Zheng | Trajectory data mining: an overview | |
| Murdukhayeva et al. | Assessment of inundation risk from sea level rise and storm surge in northeastern coastal national parks | |
| US9435657B2 (en) | Navigation system with an itinerary planning mechanism and method of operation thereof | |
| KR101724259B1 (en) | Interested area analyzing method by an integrated management system of disaster safety | |
| CN101040554B (en) | Destination prediction apparatus and destination prediction method | |
| CN102142215B (en) | Adaptive geographic information voice explanation method based on position and speed | |
| Chen et al. | Indoor cartography | |
| CN102682573A (en) | Time sequence analysis-based storm surge disaster early warning system | |
| CN115017246A (en) | Holographic navigation scene graph system for intelligent navigation of ship and construction method | |
| CN102162854A (en) | Method for acquiring position information in real time based on GPS (Global Positioning System) data | |
| US10672114B1 (en) | Computationally efficient distance-based score approximations | |
| US11385063B1 (en) | Computationally efficient distance-based score approximations | |
| US20220286804A1 (en) | Recommending targeted locations and optimal experience time | |
| US20220119223A1 (en) | Determining floor name based on audio and/or visual samples | |
| Beeharee et al. | Filtering location-based information using visibility | |
| Sueda et al. | Social geoscape: visualizing an image of the city for mobile UI using user generated geo-tagged objects | |
| Dewali et al. | A GPS-based real-time avalanche path warning and navigation system | |
| KR101181294B1 (en) | Positioning method of reported object image and processing system thereof | |
| Sun et al. | GPS data cleaning and analysis based on YouSense mobile application | |
| Huang et al. | Advanced technology in GIS/GPS-based transportation systems | |
| Cremona et al. | Speed Aware–a mobile app prototype for the promotion of responsible driving |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20121031 Termination date: 20210927 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |