CN104462727A - Oil spilling simulation parameter optimization method based on dynamic remote sensing data driving - Google Patents
Oil spilling simulation parameter optimization method based on dynamic remote sensing data driving Download PDFInfo
- Publication number
- CN104462727A CN104462727A CN201410852667.4A CN201410852667A CN104462727A CN 104462727 A CN104462727 A CN 104462727A CN 201410852667 A CN201410852667 A CN 201410852667A CN 104462727 A CN104462727 A CN 104462727A
- Authority
- CN
- China
- Prior art keywords
- oil spilling
- ecom
- oil
- simulation
- parameter
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an oil spilling simulation parameter optimization method based on dynamic remote sensing data driving. The method includes the following steps of 1, the oil spilling information remote sensing retrieval process, 2, the ECOM oil spilling simulation process, 3, the comparison and verification process of a remote sensing retrieval result and an ECOM simulation result, 4, the data integration and optimization process, and 5, the system feedback and parameter optimization process. The method has the advantages that the dynamic data driving thought is introduced, an oil spilling simulation result is verified and optimized through remote sensing retrieval oil spilling information, oil spilling simulation initial parameters are optimized through the BP neural network model parameter training and predication processes, a dynamic feedback adjusting system is formed by remote sensing monitoring and oil spilling simulation, and the accurate oil spilling dispersion motion trend is acquired through predication.
Description
Technical field
The present invention relates to the oil spilling emulation technology of dynamic remote data-driven, specifically, relate to a kind of oil spilling parameter optimization method based on dynamic remote data-driven.
Background technology
On June 4th, 2011 and 17 days, successively there are two oil spill accidents in 19-3 oil field, Peng Lai; 19-3 oil field, Peng Lai oil spill accident is the first extensive sub-sea drilled wells oil spill events in ground in China in Recent Years.According to Co., Ltd's statistics in Kang Fei oil, total about 700 barrels of crude oil leak out to sea, the Bohai Sea, and 2,500 barrels of mineral oil oil-base mud seepages of separately having an appointment also deposit to sea bed.National Bureau of Oceanography represents, current accident causes 5, and 500 square kilometres of seawater are contaminated, are roughly equivalent to 7% of Bohai Sea area.
In view of marine oil spill accident frequently occurs, causing serious harm to public waters environment and public safety, obtain oil spilling information, prediction oil spilling movement locus accurately and rapidly, is that we need the direction of exploratory development badly.Because remote sensing technology has the feature such as broad covered area, economy, easily quick obtaining, it is the effective means of marine oil spill pollution monitoring.Utilize remote sensing satellite data not only can the large area monitoring area of marine oil overflow, kind and thickness, timely guiding maritime patrol ship and aircraft carry out law enforcement monitoring, but also satellite Continuous Tracking greasy dirt scope and oil spilling dispersal direction can be utilized, formulate best Treatment scheme.Wherein, satellite-borne synthetic aperture radar (SAR) has that round-the-clock, round-the-clock, area coverage are large, quick obtaining and close to the feature such as real-time, utilizing SAR remote sensing technology can monitor marine oil spill in time, accurately, all sidedly to pollute, is the most effective means of spilled oil monitoring.
For the simulation of oil spilling movement locus, widely used is at present ECOM(Estuarine Coastal and Ocean Model) pattern.ECOM pattern is at POM(Princeton Ocean Model) basis grows up one comparatively ripe shallow sea Three-dimensional Hydrodynamic pattern.It is from initial three-dimensional equation, using free water elevation, three direction speed components, temperature, salinity, density and two characteristic quantities representing turbulent flow: tubulence energy, rapid macro-scale, as prediction variable, can obtain the forecast result of the level of step-length continuous time, the meticulous of vertical direction.But marine oil overflow accident has complicated generation, development and evolution mechanism, also there is secondary disaster and disaster coupled problem simultaneously; By the impact of the features such as mission nonlinear, time variation, multivariate and uncertainty, be difficult to by behavior complicated in traditional analytic model and realistic model analysis and prediction oil spill accident emergency system; And current realistic model mainly relies on history casualty data and empirical hypothesis initial parameter to carry out simulation analysis, the real-time observed data in emergency system can not be utilized to improve the accuracy of realistic model prediction.
For the problem in correlation technique, at present effective solution is not yet proposed.
Summary of the invention
The object of this invention is to provide a kind of oil spilling parameter optimization method based on dynamic remote data-driven, introduce dynamic data driven application system, the detection of remote sensing oil spilling is emulated with oil spilling and combines, by the method for real system data-mapping to realistic model, the update algorithm and oil spill accident modeling algorithm etc. of realistic model state, object improves the forecasting accuracy of oil spilling realistic model, improves the science of Emergency decision.Dynamic data driven application system (Data Driven Application System, DDDAS) be a kind of novel research mode that test or measurement data and emulation can be closely linked, by additional data dynamically being introduced model when simulation run, and abstract, refinement initial model or carry out model replacement selectively on this basis, the dynamic feedback between realistic model and experiment can be realized, effectively overcome currently available technology above shortcomings.
The object of the invention is to be achieved through the following technical solutions:
Based on an oil spilling parameter optimization method for dynamic remote data-driven, comprise the following steps:
Step 1: the feature being different from seawater according to the marine oil spill of the SAR image gathered in advance, utilizes Threshold Segmentation Algorithm, extracts the minimum atural object of rank as offshore spilled oil information by image greyscale partition of the level;
Step 2: according to the oil spilling history casualty data gathered in advance, set up ECOM realistic model, and arrange initial parameter to ECOM realistic model, starts ECOM realistic model, obtains emulation oil film information;
Step 3: the ECOM simulation result that the offshore spilled oil information of Remotely sensed acquisition step 1 obtained and step 2 obtain carries out contrast verification, judges whether ECOM oil spilling simulation result meets the standard information pre-set; Under described ECOM oil spilling simulation result with the incongruent situation of standard information that pre-sets, remote-sensing inversion oil spilling information and ECOM Simulation result data are integrated and are optimized, wherein, described in meet and comprise identical or conform to;
Step 4: remote-sensing inversion oil spilling information and ECOM simulation result are carried out Data Integration and optimization, the ECOM oil spilling simulation result be optimized;
Step 5: the oil spilling initial parameter of step 2 and emulation oil film information are carried out training BP neural network, and the BP neural network of the ECOM oil spilling simulation result of the optimization that step 4 is obtained input training, utilize the information prediction of remote sensing oil spilling inverting oil spilling to draw the initial parameter of oil spilling, and the initial parameter of ECOM realistic model is optimized.
Further, in step 3, when ECOM oil spilling simulation result meets the standard pre-set, offshore spilled oil motion simulation is terminated, turns back to step 2, enter into next ECOM oil spilling simulation process.
Further, described step 5 comprises:
Step 5-1:BP neural network model training process, select the input of any ECOM emulation, output parameter as the input of BP neural network, output parameter, and then training draws the input of ECOM realistic model, the Nonlinear Mapping relation of output parameter;
Step 5-2:BP Neural Network model predictive process, the BP neural network drawn is trained according to step 5-1, select remote-sensing inversion and ECOM simulation comparison to verify the input parameter of oil spilling information as neural network of the optimization drawn, drawn the oil spilling simulation parameter of optimization by the prediction of ECOM realistic model.
Beneficial effect of the present invention is: by introducing the thought that dynamic data drives, the Information Authentication of remote-sensing inversion oil spilling is utilized to optimize oil spilling simulation result, and realize the optimization of oil spilling emulation initial parameter by BP neural network model parameter training and forecasting process, remote sensing monitoring and oil spilling is made to emulate formation dynamic looped system, effectively utilize the dynamic remote data obtained in real time, the initial parameter of ECOM oil spilling emulation is optimized, makes remote sensing monitoring and oil spilling emulate formation dynamic feedback control system; Reduce the error that traditional simulation process is caused by empirical parameter, can after oil spill accident occur, draw a circle to approve greasy dirt position rapidly and predict its movement tendency, the emergency command controlled for greasy dirt provides decision information, thus prediction obtains comparatively accurate oil spilling diffusion motion trend.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the process flow diagram of a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention;
Fig. 2 is the system feedback of dynamic remote data-driven and the process flow diagram of parameter optimisation procedure of a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention;
Fig. 3 is the grey level histogram of the ASAR image that on June 11st, 2011 of a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention obtains;
Fig. 4 is the grey level histogram of the ASAR image that on June 14th, 2011 of a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention obtains;
Fig. 5 is Peng Lai 19-3B platform oil spilling extraction effect figure of the ASAR image that on June 11st, 2011 of a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention obtains;
Fig. 6 is Peng Lai 19-3B platform oil spilling extraction effect figure of the ASAR image that on June 14th, 2011 of a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention obtains;
Fig. 7 is Peng Lai 19-3B platform oil spilling elaioleucite movement locus (part) schematic diagram one that the oil spilling simulation parameter of utilization optimization according to the embodiment of the present invention is simulated continuous 15 days that draw;
Fig. 8 is Peng Lai 19-3B platform oil spilling elaioleucite movement locus (part) schematic diagram two that the oil spilling simulation parameter of utilization optimization according to the embodiment of the present invention is simulated continuous 15 days that draw;
Fig. 9 is Peng Lai 19-3B platform oil spilling elaioleucite movement locus (part) schematic diagram three that the oil spilling simulation parameter of utilization optimization according to the embodiment of the present invention is simulated continuous 15 days that draw;
Figure 10 is Peng Lai 19-3B platform oil spilling elaioleucite movement locus (part) schematic diagram four that the oil spilling simulation parameter of utilization optimization according to the embodiment of the present invention is simulated continuous 15 days that draw;
Figure 11 is Peng Lai 19-3B platform oil spilling elaioleucite movement locus (part) schematic diagram five that the oil spilling simulation parameter of utilization optimization according to the embodiment of the present invention is simulated continuous 15 days that draw;
Figure 12 is Peng Lai 19-3B platform oil spilling elaioleucite movement locus (part) schematic diagram six that the oil spilling simulation parameter of utilization optimization according to the embodiment of the present invention is simulated continuous 15 days that draw;
Figure 13 is two days remote-sensing inversion oil spilling information that there is remotely-sensed data on June in 2011 11 according to the embodiment of the present invention and the superimposed figure of ECOM simulation result;
Figure 14 is two days remote-sensing inversion oil spilling information that there is remotely-sensed data on June in 2011 14 according to the embodiment of the present invention and the superimposed figure of ECOM simulation result;
Figure 15 is that remote-sensing inversion according to the embodiment of the present invention and ECOM emulate elaioleucite angle schematic diagram.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain, all belongs to the scope of protection of the invention.
In view of the more difficult prediction of marine oil spill accident development trend taken place frequently in recent years, introduce the thought that dynamic data drives, remote sensing spilled oil monitoring, oil spilling simulation numerical model are combined with geographic information system technology, invent a kind of oil spilling simulation parameter optimization method based on dynamic remote data-driven, set up with this aggregation of data treatment and analyses prediction emergency reaction integrated system driven based on dynamic data.Oil spill accident is once occur, the oil spilling information utilizing remote sensing image to extract will as the precision evaluation standard of oil spilling emulation and driven factor, by feedback regulation Optimized model parameter, finally obtain comparatively accurate continuous print spilled oil simulation result, provide foundation and technical support as the formulation of the analysis of oil spill accident, emergency reaction scheme and Disaster Assessment etc.
As shown in figures 1-15, a kind of oil spilling parameter optimization method based on dynamic remote data-driven according to the embodiment of the present invention; Specifically comprise the following steps:
Step 1: oil spilling information remote sensing refutation process, is different from the feature of seawater, utilizes Threshold Segmentation Algorithm according to the marine oil spill of the SAR image gathered in advance, extract the minimum atural object of rank as offshore spilled oil information by image greyscale partition of the level;
Step 2:ECOM oil spilling simulation process, according to the oil spilling history casualty data gathered in advance, arranges ECOM realistic model, and arranges initial parameter to ECOM realistic model, starts ECOM realistic model, obtains emulation oil film information;
Step 3: remote-sensing inversion result and ECOM simulation result contrast verification process, the ECOM simulation result that the offshore spilled oil information of Remotely sensed acquisition step 1 obtained and step 2 obtain carries out contrast verification, judges whether ECOM oil spilling simulation result meets the standard pre-set; Under described ECOM oil spilling simulation result with the incongruent situation of standard information that pre-sets, remote-sensing inversion oil spilling information and ECOM Simulation result data are integrated and are optimized, wherein, described in meet and comprise identical or conform to; Conform to the data error scope pre-set, as long as described ECOM oil spilling simulation result is in the scope pre-set, is and meets.
Step 4: Data Integration and optimizing process, utilizes prior art to carry out integration and the optimization of data remote-sensing inversion oil spilling information and ECOM simulation result, the ECOM oil spilling simulation result be optimized;
Step 5: system feedback and parameter optimisation procedure, first the oil spilling initial parameter of step 2 and emulation oil film information are carried out training BP neural network, the BP neural network of the ECOM oil spilling simulation result input training of the optimization then step 4 obtained, utilize the information prediction of remote sensing oil spilling inverting oil spilling to draw the initial parameter of oil spilling, and the initial parameter of realistic model is optimized.
In step 3, when ECOM oil spilling simulation result meets the standard pre-set, offshore spilled oil motion simulation is terminated, turns back to step 2, enter into next ECOM oil spilling simulation process.
Select BP neural network model, by the input/output argument training BP neural network to ECOM model, then utilize remote sensing oil spilling inverting oil spilling information prediction oil spilling initial parameter, thus the step realizing parameter optimization comprise:
Step 5-1:BP neural network model training process, select the input of any ECOM emulation, output parameter as the input of BP neural network, output parameter, and then training draws the input of ECOM realistic model, the Nonlinear Mapping relation of output parameter;
Step 5-2:BP Neural Network model predictive process, the BP neural network drawn is trained according to step 5-1, select remote-sensing inversion and ECOM simulation comparison to verify the input parameter of oil spilling information as neural network of the optimization drawn, drawn the oil spilling simulation parameter of optimization by the prediction of ECOM realistic model.
Because maritime patrol ship in marine oil spill accident is difficult to the information such as the oil spill area accurately obtaining sea at short notice, remote sensing technology is due to wide, quick, the economic feature of area coverage, the oil spilling information on sea can be obtained at short notice, and in this, as the contrast verification standard of simulation result.Utilize dynamic remote inverting information contrast verification and adjust ECOM simulation result, by contrast verification, the ECOM that can be optimized emulates oil spilling result; Therefore, the present invention can the area, position etc. of large area monitoring marine oil overflow, guides in time maritime patrol ship and aircraft to carry out law enforcement and monitors.
Due to the nonlinear characteristic of ECOM model, contrast the anti-starting condition pushing away realistic model of the oil spilling diffuseness values optimizing the renewal drawn by remote-sensing inversion oil spilling information and ECOM simulation result and there is difficulty greatly, we select BP neural network model, by the input/output argument training BP neural network to ECOM model, then utilize remote sensing oil spilling inverting oil spilling information prediction oil spilling initial parameter, thus realize parameter optimization.
During concrete operations, for step 1, oil spilling information remote sensing refutation process: " secretly " feature being different from seawater according to SAR image marine oil spill, utilizes Threshold Segmentation Algorithm, extracts the minimum atural object of rank as offshore spilled oil information by image greyscale partition of the level; Wherein, the object of Iamge Segmentation is that separation of images is become significant region not overlapping mutually, and the partitioning algorithm of monochrome image is usually based on uncontinuity and the similarity of image brightness values, and single Threshold Segmentation Algorithm is selected in this research.
The technical program using Peng Lai, the Bohai Sea in June, 2011 19-3B oil spill accident as experiment case study on implementation, the ENVISAT ASAR WS mode data selecting on June 11st, 2011,14 days to obtain is as Peng Lai, Bohai Sea 19-3 oil overflowing remote sense Monitoring Data, through process such as geometric accurate correction, enhanced Lee filtering, cuttings, obtain the subregion of 800*800, this Peng Lai, region overlay Bohai Sea 19-3 B, C production vessel.
Due to the ASAR image containing oil spilling phenomenon, its grey level histogram generally shows as existence two peak values, can choose two peak-to-peak minimum value as threshold value to distinguish oil film and background seawater region.By statistics of histogram, the grey level histogram of two width remote sensing images can be drawn respectively as shown in Figure 3-4; Can be drawn by Fig. 3, June 11 there are two obvious crest peak A and B in ASAR image, and therefore, we choose the segmentation threshold of the peak-to-peak minimum gray value 526 of two ripples as the information extraction of remote sensing oil spilling; And for image on June 14 (Fig. 4), because oil spilling region area is relative to less view picture image, the peak value that its gray scale is less is not obvious, and grey level histogram only exists an obvious crest peak B, and we cannot utilize the peak-to-peak minimum gradation value of two ripples to determine segmentation threshold.But " secretly " feature presented in SAR image due to oil film overlay area, there are two obvious flex points respectively in its grey level histogram 174 and 216, we choose 174,216 respectively as segmentation threshold, compare through experiment, draw 216 segmentation thresholds that can be used as ASAR image oil spilling information extraction on June 14.
Through June 11,14 days oil spilling information extraction result that single Threshold Segmentation Algorithm obtains as seen in figs. 5-6.Carry out classification accuracy assessment by choosing oil film sample at original ASAR image, it is 0.9425 that Threshold Segmentation Algorithm extraction oil film precision can be utilized to be 92.1726%, Kappa coefficient, meets the standard as next step oil spilling movement locus simplation verification.
For step 2, ECOM oil spilling simulation process: according to the history oil spill accident data gathered in advance and expertise, suppose the initial parameter (table 1) such as Position Approximate, time, tonnage showing that Peng Lai 19-3B oil spill accident occurs, start ECOM realistic model, obtain first simulation result;
The oil spilling emulation initial parameter that table 1 is supposed
| Oil spilling point X | Oil spilling point Y | Step interval | Elaioleucite release number | Elaioleucite release starting step size | Elaioleucite release terminates step-length |
| 93 | 52 | 30 | 100 | 961 | 7860 |
For step 3, whether remote-sensing inversion result and ECOM simulation result contrast verification process: the oil spilling information of the Remotely sensed acquisition utilizing step 1 to obtain carries out contrast verification to the ECOM simulation result that step 2 obtains, evaluate ECOM oil spilling simulation result and meet.If meet, then the oil spilling motion simulation of this moment terminates, and can enter subsequent time simulation, i.e. step 2; If do not meet, then enter remote-sensing inversion oil spilling information and ECOM Simulation result data is integrated and optimizing process, i.e. step 4.
For step 4, Data Integration and optimizing process: select remote-sensing inversion oil spilling information and ECOM simulation result to carry out Data Integration and optimization, the ECOM oil spilling simulation result be optimized.
For step 5, system feedback and parameter optimisation procedure: the oil spilling initial parameter and the simulation result training BP neural network that first utilize step 2, the BP neural network of the oil spilling simulation result input training of the optimization then step 4 obtained, pass through feedback forecasting, anti-release oil spilling starting condition, and the initial parameter of realistic model is optimized (table 2).
The oil spilling emulation initial parameter that table 2 is optimized
| Oil spilling point X | Oil spilling point Y | Step interval | Elaioleucite release number | Elaioleucite release starting step size | Elaioleucite release terminates step-length |
| 91 | 53 | 50 | 80 | 961 | 4320 |
According to above introduction, design completes a set of oil spilling simulation parameter optimization method based on dynamic remote data-driven.This method given full play to remote sensing in real time, the advantage of fast monitored large area offshore spilled oil, improve the precision of oil spilling emulation, well reduce the error that traditional simulation is caused by empirical parameter.
In order to verify the inventive method performance, the oil spilling simulation parameter simulation optimized is utilized to draw Peng Lai 19-3B oil spill accident elaioleucite movement locus (Fig. 7-12) of continuous 15 days, and select exist remotely-sensed data within 2 days, carry out the information inverting of remote sensing oil spilling and accuracy comparison checking (Figure 13-14), concrete grammar is as follows:
ECOM realistic model can obtain the elaioleucite movement locus of continuous a period of time, calculates the central point longitude and latitude of each moment elaioleucite respectively, is then coupled together by these central points, can obtain ECOM and emulate elaioleucite motion vector; In like manner, we also can draw the motion vector of remote-sensing inversion elaioleucite; According to these two motion vectors, we can calculate the emulation of ECOM oil spilling and the dynamic diffusion tendency coupling efficiency of remote-sensing inversion elaioleucite, draw the precision of realistic model with this.
The generalized angle cosine defined between two motion vectors is similar function, and wherein, X and Y represents longitude and latitude respectively, uses for reference cosine function theorem, and we think angle >=90 degree, then be completely reverse, spreading accuracy is 0; 0 degree for be coupled completely, diffusion accuracy be 100%; Diffusion accuracy between 0-90 degree is the cosine value of generalized angle.Angle is less, shows that it is more similar, and namely to emulate the dynamic diffusion tendency obtained more similar for the dynamic diffusion tendency of the elaioleucite of remote sensing monitoring and oil spilling.The cosine of two vector generalized angle is:
（1）
Selection 20110611,20110614 exists two days of remotely-sensed data, first calculates this emulation elaioleucite of two days, Remotely sensed acquisition elaioleucite center longitude, then calculates the angle (Figure 15) of two vector line segments that these four central points are formed.In this experiment,
（2）
Then
, namely the dynamic diffusion tendency coupling efficiency of elaioleucite is 99.97%.
In sum, by means of technique scheme of the present invention, by using remote-sensing inversion oil spilling information as precision evaluation standard, dynamic data feedback regulation and update algorithm are incorporated in oil spilling parameter optimisation procedure, set up with the BP neural network parameter training that is core, prediction and Optimized model, the initial parameter that oil spilling is emulated obtains optimum solution in the Nonlinear Mapping of training; Thus the error that the empirical hypothesis parameter avoiding traditional simulation model is brought, can obtain comparatively accurate oil spilling simulation result.
In addition, the thought that the technical program drives by introducing dynamic data, the Information Authentication of remote-sensing inversion oil spilling is utilized to optimize oil spilling simulation result, and realize the optimization of oil spilling emulation initial parameter by BP neural network model parameter training and forecasting process, remote sensing monitoring and oil spilling is made to emulate formation dynamic looped system, effectively utilize the dynamic remote data obtained in real time, the initial parameter of ECOM oil spilling emulation is optimized, makes remote sensing monitoring and oil spilling emulate formation dynamic feedback control system; Reduce the error that traditional simulation process is caused by empirical parameter, can after oil spill accident occur, draw a circle to approve greasy dirt position rapidly and predict its movement tendency, the emergency command controlled for greasy dirt provides decision information, thus prediction obtains comparatively accurate oil spilling diffusion motion trend.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (3)
1., based on an oil spilling simulation parameter optimization method for dynamic remote data-driven, it is characterized in that, comprise the following steps:
Step 1: the feature being different from seawater according to the marine oil spill of the SAR image gathered in advance, utilizes Threshold Segmentation Algorithm, extracts the minimum atural object of rank as offshore spilled oil information by image greyscale partition of the level;
Step 2: according to the oil spilling history casualty data gathered in advance, set up ECOM realistic model, and arrange initial parameter to ECOM realistic model, starts ECOM realistic model, obtains emulation oil film information;
Step 3: the ECOM simulation result that the offshore spilled oil information of Remotely sensed acquisition step 1 obtained and step 2 obtain carries out contrast verification, judges whether ECOM oil spilling simulation result meets the standard information pre-set; Under described ECOM oil spilling simulation result with the incongruent situation of standard information that pre-sets, remote-sensing inversion oil spilling information and ECOM Simulation result data are integrated and are optimized, wherein, described in meet and comprise identical or conform to;
Step 4: remote-sensing inversion oil spilling information and ECOM simulation result are carried out Data Integration and optimization, the ECOM oil spilling simulation result be optimized;
Step 5: the oil spilling initial parameter of step 2 and emulation oil film information are carried out training BP neural network, and the BP neural network of the ECOM oil spilling simulation result of the optimization that step 4 is obtained input training, utilize the information prediction of remote sensing oil spilling inverting oil spilling to draw the initial parameter of oil spilling, and the initial parameter of ECOM realistic model is optimized.
2. the oil spilling simulation parameter optimization method based on dynamic remote data-driven according to claim 1, it is characterized in that, in step 3, when ECOM oil spilling simulation result meets the standard pre-set, offshore spilled oil motion simulation is terminated, turn back to step 2, enter into next ECOM oil spilling simulation process.
3. the oil spilling simulation parameter optimization method based on dynamic remote data-driven according to claim 1, it is characterized in that, described step 5 comprises:
Step 5-1:BP neural network model training process, select the input of any ECOM emulation, output parameter as the input of BP neural network, output parameter, and then training draws the input of ECOM realistic model, the Nonlinear Mapping relation of output parameter;
Step 5-2:BP Neural Network model predictive process, the BP neural network drawn is trained according to step 5-1, select remote-sensing inversion and ECOM simulation comparison to verify the input parameter of oil spilling information as neural network of the optimization drawn, drawn the oil spilling simulation parameter of optimization by the prediction of ECOM realistic model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410852667.4A CN104462727B (en) | 2014-12-31 | 2014-12-31 | A kind of oil spilling simulation parameter optimization method based on dynamic remote data-driven |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410852667.4A CN104462727B (en) | 2014-12-31 | 2014-12-31 | A kind of oil spilling simulation parameter optimization method based on dynamic remote data-driven |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104462727A true CN104462727A (en) | 2015-03-25 |
| CN104462727B CN104462727B (en) | 2018-08-24 |
Family
ID=52908757
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410852667.4A Active CN104462727B (en) | 2014-12-31 | 2014-12-31 | A kind of oil spilling simulation parameter optimization method based on dynamic remote data-driven |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104462727B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107807069A (en) * | 2017-10-25 | 2018-03-16 | 中国石油大学(华东) | The adaptive tracking control method and its system of a kind of offshore spilled oil |
| CN108564193A (en) * | 2018-01-05 | 2018-09-21 | 广东电网有限责任公司中山供电局 | The prediction technique of electric power personal injury, misoperation fault and the accident of operation against rules against regulations based on neural network |
| CN108596065A (en) * | 2018-04-13 | 2018-09-28 | 深圳职业技术学院 | One kind is based on deep semantic segmentation marine oil spill detecting system and method |
| CN109784320A (en) * | 2019-03-25 | 2019-05-21 | 中国科学院地理科学与资源研究所 | Ginseng industrialized agriculture domain determines method |
| CN110485222A (en) * | 2019-07-29 | 2019-11-22 | 中国铁路总公司 | A kind of dynamic data inversion method and device |
| CN111257854A (en) * | 2020-01-19 | 2020-06-09 | 中南林业科技大学 | Universal terrain correction optimization method based on remote sensing image segmentation unit |
| CN111680889A (en) * | 2020-05-20 | 2020-09-18 | 中国地质大学（武汉） | Offshore oil leakage source positioning method and device based on cross entropy |
| CN116562186A (en) * | 2023-05-10 | 2023-08-08 | 南京工程学院 | Underwater gas transmission pipeline leakage parameter inversion method and system based on simulation-optimization |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102508946A (en) * | 2011-10-10 | 2012-06-20 | 江苏科技大学 | Method for simulating spilled oil sea surface under finite water depth |
-
2014
- 2014-12-31 CN CN201410852667.4A patent/CN104462727B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102508946A (en) * | 2011-10-10 | 2012-06-20 | 江苏科技大学 | Method for simulating spilled oil sea surface under finite water depth |
Non-Patent Citations (4)
| Title |
|---|
| OSCAR GARCIA-PINEDA ET AL: "Oil Spill Mapping and Measurement in the Gulf of Mexico With Textural Classifier Neural Network Algorithm(TCNNA)", 《IEEE JOURNAL OF SELECTED IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING》 * |
| YAO LI ET AL: "Application of DDDAS in marine oil spill management: A new framework combining multiple source remote sensing monitoring and simulation as a symbiotic feedback control system", 《IGARSS》 * |
| YONGGANG LIU ET AL: "Trajectory Forecast as a Rapid Response to the Deepwater Horizon Oil Spill", 《MONITORING AND MODELING THE DEEPWATER HORIZON OIL SPILL: A RECORD-BREAKING ENTERPRISE》 * |
| 郭良波 等: "番禺4-2油田溢油数值模拟", 《海洋湖沼通报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107807069A (en) * | 2017-10-25 | 2018-03-16 | 中国石油大学(华东) | The adaptive tracking control method and its system of a kind of offshore spilled oil |
| CN108564193A (en) * | 2018-01-05 | 2018-09-21 | 广东电网有限责任公司中山供电局 | The prediction technique of electric power personal injury, misoperation fault and the accident of operation against rules against regulations based on neural network |
| CN108596065A (en) * | 2018-04-13 | 2018-09-28 | 深圳职业技术学院 | One kind is based on deep semantic segmentation marine oil spill detecting system and method |
| CN109784320A (en) * | 2019-03-25 | 2019-05-21 | 中国科学院地理科学与资源研究所 | Ginseng industrialized agriculture domain determines method |
| CN110485222A (en) * | 2019-07-29 | 2019-11-22 | 中国铁路总公司 | A kind of dynamic data inversion method and device |
| CN110485222B (en) * | 2019-07-29 | 2020-11-27 | 中国铁路总公司 | Dynamic data inversion method and device |
| CN111257854A (en) * | 2020-01-19 | 2020-06-09 | 中南林业科技大学 | Universal terrain correction optimization method based on remote sensing image segmentation unit |
| CN111680889A (en) * | 2020-05-20 | 2020-09-18 | 中国地质大学（武汉） | Offshore oil leakage source positioning method and device based on cross entropy |
| CN111680889B (en) * | 2020-05-20 | 2023-08-18 | 中国地质大学（武汉） | Cross entropy-based offshore oil leakage source positioning method and device |
| CN116562186A (en) * | 2023-05-10 | 2023-08-08 | 南京工程学院 | Underwater gas transmission pipeline leakage parameter inversion method and system based on simulation-optimization |
| CN116562186B (en) * | 2023-05-10 | 2023-12-26 | 南京工程学院 | Underwater gas transmission pipeline leakage parameter inversion method and system based on simulation-optimization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104462727B (en) | 2018-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104462727A (en) | Oil spilling simulation parameter optimization method based on dynamic remote sensing data driving | |
| Zamani et al. | Learning from data for wind–wave forecasting | |
| Xu et al. | Oil spill forecast model based on uncertainty analysis: A case study of Dalian Oil Spill | |
| Liu et al. | Assessing pollution-related effects of oil spills from ships in the Chinese Bohai Sea | |
| Goharnejad et al. | Vulnerability assessment of southern coastal areas of Iran to sea level rise: evaluation of climate change impact | |
| Aydog et al. | Current velocity forecasting in straits with artificial neural networks, a case study: Strait of Istanbul | |
| Zhang et al. | Analysis of 50-year wind data of the southern Baltic Sea for modelling coastal morphological evolution–a case study from the Darss-Zingst Peninsula | |
| Chiri et al. | Mid-long term oil spill forecast based on logistic regression modelling of met-ocean forcings | |
| Yang et al. | An approach to ship behavior prediction based on AIS and RNN optimization model | |
| Chen et al. | Offshore oil spill monitoring and detection: Improving risk management for offshore petroleum cyber-physical systems | |
| Kisi et al. | Forecasting sea water levels at Mukho Station, South Korea using soft computing techniques | |
| Wang et al. | A dynamic marine oil spill prediction model based on deep learning | |
| Sembiring et al. | Nearshore bathymetry from video and the application to rip current predictions for the Dutch Coast | |
| Garrity et al. | Evaluation of event and response approaches to estimate the 100-year coastal flood for Pacific coast sheltered waters | |
| Quy et al. | Risk-and simulation-based optimization of channel depths: Entrance channel of Cam Pha Coal Port | |
| Guo et al. | Numerical modelling of temporal and spatial patterns of petroleum hydrocarbons concentration in the Bohai Sea | |
| Yu et al. | i4OilSpill, an operational marine oil spill forecasting model for Bohai Sea | |
| Xiang et al. | Simulation of Remote-Sensed Chlorophyll Concentration with a Coupling Model Based on Numerical Method and CA-SVM in Bohai Bay, China | |
| Yao et al. | Extreme motion predictions of semi-submersible platform based on deep learning method and prototype monitoring data | |
| Vanem et al. | Comparison of the environmental contour method and response-based analysis using response emulator for estimating extreme ship responses | |
| Toulemonde et al. | Space–Time Simulations of Extreme Rainfall: Why and How? | |
| CROWLEY et al. | SHORELINE CHANGE IMPACTS ON COASTAL VULNERABILITY TO NOREASTERS | |
| CN115019168B (en) | SAR image water body extraction method and system based on mixed attention network | |
| Jaouën et al. | Methodology for the Design of LNG Terminals in a Nearshore Environment | |
| Van den Eynde et al. | The impact of sand extraction on the wave height near the Belgian coast |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |