CN117576951B - Cross-sea bridge active early warning method based on ship collision risk probability identification - Google Patents
Cross-sea bridge active early warning method based on ship collision risk probability identification Download PDFInfo
- Publication number
- CN117576951B CN117576951B CN202410060780.2A CN202410060780A CN117576951B CN 117576951 B CN117576951 B CN 117576951B CN 202410060780 A CN202410060780 A CN 202410060780A CN 117576951 B CN117576951 B CN 117576951B
- Authority
- CN
- China
- Prior art keywords
- ship
- bridge
- risk
- early warning
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 52
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 230000001960 triggered effect Effects 0.000 claims description 17
- 238000005457 optimization Methods 0.000 claims description 11
- 239000003643 water by type Substances 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 239000000463 material Substances 0.000 claims description 3
- 238000004880 explosion Methods 0.000 claims description 2
- 238000011160 research Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000002457 bidirectional effect Effects 0.000 description 4
- 230000006378 damage Effects 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 238000009863 impact test Methods 0.000 description 3
- 238000012502 risk assessment Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 208000027418 Wounds and injury Diseases 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 208000014674 injury Diseases 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012916 structural analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G3/00—Traffic control systems for marine craft
- G08G3/02—Anti-collision systems
Landscapes
- Engineering & Computer Science (AREA)
- Ocean & Marine Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The application discloses a cross-sea bridge active early warning method based on ship collision risk probability identification, which comprises the following steps: firstly, acquiring a bridge collapse probability table under the collision of a ship bridge; predicting the track of the ship, and calculating the probability of the ship striking the bridge according to the predicted trackA striking position; inquiring a bridge collapse probability table based on the tonnage, the navigational speed and the predicted impact position of the ship to obtain the bridge collapse probability corresponding to the impact positionThe method comprises the steps of carrying out a first treatment on the surface of the Probability of risk of ship collisionThe method comprises the steps of carrying out a first treatment on the surface of the Finally according to the calculated ship collision risk probabilityAnd the distance between the ship and the bridge sends a corresponding early warning signal. The beneficial effects of this application: the ship collision risk probability provided by the application covers the bridge collapse probability, the ship collision bridge probability and the water area risk level, the ship collision bridge risk research points in the three aspects are comprehensively considered, quantitative calculation is carried out on the ship collision bridge risk, and a finer means is provided for active early warning of the bridge.
Description
Technical Field
The application relates to the technical field of bridge safety management, in particular to a cross-sea bridge active early warning method based on ship collision risk probability identification.
Background
In recent years, along with the continuous perfection of infrastructure, the number of bridges crossing the river and the sea is continuously increased, and the bridge crossing method brings a certain obstacle to the passing of ships while facilitating land traffic, so that the bridge collision accident of the ships occurs. Therefore, a great deal of research work is performed on risk analysis of ship bridge collision by students at home and abroad. Overall, the risk analysis study of ship-to-bridge includes three main aspects: firstly, analyzing the consequences of a ship bumping bridge; secondly, analyzing the probability of the ship bumping bridge; thirdly, researching risk criteria of ship bumping bridge.
At present, the application of the risk analysis of the ship bridge collision in the bridge active early warning method mainly considers the second aspect, namely the probability analysis of the ship bridge collision: the probability of the ship bumping bridge is calculated by adopting the data of equipment such as AIS, radar, camera and the like and adopting the means of big data and track prediction so as to achieve the purpose of early warning the risk ship. Therefore, although the effect of active early warning can be realized to a certain extent, the ship collision risk considered by the method is incomplete, and the accurate identification of the ship collision risk probability cannot be realized.
Disclosure of Invention
One of the purposes of the application is to provide a cross-sea bridge active early warning method based on ship collision risk probability identification, which can solve at least one defect in the background art.
In order to achieve at least one of the above objects, the technical scheme adopted in the application is as follows: a cross-sea bridge active early warning method based on ship collision risk probability identification comprises the following steps:
s100: acquiring a bridge collapse probability table under the collision of the ship bridge;
s200: predicting the track of the ship, and calculating the probability of the ship striking the bridge according to the predicted trackA striking position;
s300: inquiring a bridge collapse probability table based on the tonnage, the navigational speed and the predicted impact position of the ship to obtain the bridge collapse probability corresponding to the impact positionThe method comprises the steps of carrying out a first treatment on the surface of the The risk probability of ship collision->；
S400: according to the calculated ship collision risk probabilityAnd the distance between the ship and the bridge sends a corresponding early warning signal.
Preferably, in step S100, the acquisition of the bridge collapse probability table includes the following processes:
s110: obtaining structural parameters of a target bridge;
s120: the method comprises the steps of statistically analyzing typical ships in the bridge site water area, and dividing the ships into different tonnage levels according to different types;
s130: building a finite element model of the bridge, performing impact analysis on different parts of the bridge on ships with different tonnages and navigational speeds, and outputting the corresponding probability of collapse of the bridge；
S140: probability of collapse of the bridge to be obtainedAnd drawing a bridge collapse probability table according to the tonnage, the navigational speed and the impact position of the ship.
Preferably, in step S110, the structural parameters of the target bridge include geometric parameters, material properties, and boundary conditions.
Preferably, in step S200, a trajectory prediction algorithm (BO-BiLSTM) combining a bidirectional long-short-term memory network (BiLSTM) and bayesian optimization (Bayesian Optimization, BO) is used to predict the trajectory of the ship, and the predicted trajectory of the ship within 10-30 minutes in the future is outputted to calculate the probability of the ship striking the bridge。
Preferably, the specific ship track prediction comprises the following processes:
s210: inputting a track sample of a ship, and building and training a BiLSTM track prediction model;
s220: taking the average value of the root mean square error of the test sample as an objective function of the BiLSTM track prediction model;
s230: performing super-parameter adjustment of the BiLSTM track prediction model by adopting Bayesian optimization;
s240: and outputting the optimized BO-BiLSTM model by taking the super parameter with the minimum objective function value as the final parameter, and outputting the predicted track of the ship within 10-30 minutes based on the BO-BiLSTM model.
Preferably, in step S210, the ship position detected by the radar system on the bridge and the position transmitted by the ship' S own AIS are combined as a track sample of the ship.
Preferably, in step S200, the probability of the ship striking the bridgeThe calculation of (1) comprises the following steps:
s250: different waters are defined according to the characteristics of the bridge area water channel and the distance from the bridge axis;
s260: acquiring probability of ship striking bridge pier based on predicted track of ship；
S270: selecting corresponding water area risk coefficient according to current water area position of shipProbability of ship striking bridge>。
Preferably, in step S250, the water areas on both sides of the bridge axis are divided into five-level risk water areas V, IV, III, II and I from far to near; then in step S270, for the water risk factorThe values of (2) are as follows: corresponding to the risk water area VA value of 0.2, corresponding to risk waters IV +.>A value of 0.4, corresponding to risk waters III +.>A value of 0.6, corresponding to risk waters II +.>The value is 0.8, corresponding to risk area I>The value is 1.
Preferably, in step S400, the risk probability of ship collision is calculated based onCarrying out risk classification, and sending early warning signals of corresponding grades according to different risk grades and the current position of the ship in the risk water area; the risk level is divided into: extremely low, medium, high and extremely high, respectively corresponding ship collision risk probabilities +.>Has a value of [0,0.01]]、（0.01，0.05]、（0.05，0. 1]、（0.1，0.5](0.5, 1)]。
Preferably, the level of the early warning signal comprises no early warning, three-level early warning, two-level early warning and one-level early warning; wherein, no early warning is that any early warning is not triggered, three-level early warning is that a message notification is sent to the ship, two-level early warning is that a flash warning lamp, a VHF call and a message notification are sent to the ship, the primary early warning is to send directional loudspeaker, explosion flash warning lamp, VHF call and message notice to the ship; when the ship is located in the risk water area V, the early warning signals are of no early warning level; when the ship is located in the risk water area IV, the three-level early warning is triggered only when the ship is at an extremely high risk level; when the ship is located in the risk water area III, triggering a secondary early warning when the ship exceeds a medium risk level, wherein the rest risk levels are not early warned; when the ship is located in a risk water area II, triggering three-level early warning when the ship is in a medium risk level, triggering one-level early warning when the ship exceeds the medium risk level, and enabling the rest risk levels to be free of early warning; when the ship is located in the risk water area I, no early warning is generated when the ship is in the extremely low risk level, the third early warning is triggered when the ship is in the low risk level, the second early warning is triggered when the ship is in the medium risk level, and the first early warning is triggered when the ship is in the medium risk level.
Compared with the prior art, the beneficial effect of this application lies in:
(1) The ship collision risk probability provided by the application covers the bridge collapse probability, the ship collision bridge probability and the water area risk level, the ship collision bridge risk research points in the three aspects are comprehensively considered, quantitative calculation is carried out on the ship collision bridge risk, and a finer means is provided for active early warning of the bridge.
(2) The intelligent algorithm prediction and the table lookup mode are combined to obtain the collapse probability result of the ship crashing the bridge, so that the calculation accuracy is met, the system operation efficiency is greatly improved, and precious time is saved for the system to make reasonable early warning judgment in time.
Drawings
FIG. 1 is a schematic overall flow chart of the present invention.
Fig. 2 is a schematic diagram of a BiLSTM network structure according to the present invention.
Fig. 3 is a schematic diagram of a level table of early warning signals corresponding to different risk levels and different risk waters in the present invention.
Detailed Description
The present application will be further described with reference to the specific embodiments, and it should be noted that, on the premise of no conflict, new embodiments may be formed by any combination of the embodiments or technical features described below.
In the description of the present application, it should be noted that, for the azimuth terms such as terms "center", "lateral", "longitudinal", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., the azimuth and positional relationships are based on the azimuth or positional relationships shown in the drawings, it is merely for convenience of describing the present application and simplifying the description, and it is not to be construed as limiting the specific protection scope of the present application that the device or element referred to must have a specific azimuth configuration and operation, as indicated or implied.
It should be noted that the terms "first," "second," and the like in the description and in the claims of the present application are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.
The terms "comprises" and "comprising," along with any variations thereof, in the description and claims of the present application are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements that are expressly listed or inherent to such process, method, article, or apparatus.
In one preferred embodiment of the present application, as shown in fig. 1, a cross-sea bridge active early warning method based on recognition of risk probability of ship collision includes the following steps:
s100: and acquiring a bridge collapse probability table under the collision of the ship bridge.
S200: predicting the track of the ship, and calculating the probability of the ship striking the bridge according to the predicted trackAnd the impact location.
S300: inquiring a bridge collapse probability table based on the tonnage, the navigational speed and the predicted impact position of the ship to obtain the bridge collapse probability corresponding to the impact positionThe method comprises the steps of carrying out a first treatment on the surface of the The risk probability of ship collision->。
S400: according to the calculated ship collision risk probabilityAnd the distance between the ship and the bridge sends a corresponding early warning signal.
It should be noted that, when the ship collides with the bridge, the ship itself may collapse while being damaged by the collision; the collapsed bridge not only can cause personal injury to personnel on the bridge, but also can cause secondary injury to ships under the bridge. In the traditional analysis of the ship bridge collision probability, only the possibility that the ship collides with the bridge is considered, and the structural response and the corresponding collapse probability generated after the ship collides with the bridge are ignored. That is, the active early warning system in the traditional mode cannot master comprehensive information to make decisions, so that the accuracy and effectiveness of the system are reduced to a certain extent.
The embodiment calculates the collapse rate of the bridge after the bridge is impacted on the basis of the probability of the bridge impact of the traditional ship; and the probability of the ship striking the bridge can be obtained by predicting the track of the ship through an intelligent algorithm. And the risk position of the ship in the water area, the collapse probability of the bridge and the collision probability of the ship on the bridge are taken as research points of the risk of the ship on the bridge, and quantitative calculation is carried out on the risk of the ship on the bridge, so that a more refined means can be provided for the active early warning of the bridge, and the accuracy and the sensitivity of the active early warning system of the bridge can be effectively improved.
It should also be appreciated that both bridges and vessels are large structures, and that numerical analysis of the structure of a ship's bump often requires significant computational resources. When navigation density is large in a channel, real-time structural analysis of more ships with different tonnages is difficult to achieve by the active early warning system of the bridge so as to give out ship collision results, so that delay of early warning of the system can be caused, and effectiveness of the system is affected.
In the embodiment, a bridge collapse rate table based on the bridge structure at different impact positions and under different forces can be obtained in advance through a simulation calculation mode and is input into an active early warning system of the bridge. Therefore, when the active early warning system of the bridge works, only the impact position of the ship needs to be determined, the corresponding bridge collapse probability can be directly inquired from the bridge collapse probability table according to the impact position of the bridge, the real-time calculated amount of the active early warning system can be effectively reduced, and the timeliness and the effectiveness of system early warning are further guaranteed.
In addition, in order to further improve the safety of the bridge, the active early warning of the ship is based on future track prediction of the current track. I.e. the current ship has a certain distance from the bridge, and the track in a certain time in the future is predicted and adjusted in real time according to the navigation track of the current ship; and then, calculating the ship collision risk probability based on the ship predicted track, so that on one hand, the active early warning system can be ensured to have enough time to calculate the ship collision risk probability, and on the other hand, when the larger ship collision risk occurs, the bridge active early warning system can timely generate early warning signals to the ship, and the ship is ensured to have enough time to finish route adjustment so as to avoid collision with the bridge.
In this embodiment, when step S100 is performed, the acquisition of the bridge collapse probability table includes the following processes:
s110: obtaining structural parameters of a target bridge; specifically, the structural parameters of the target bridge include geometric parameters, material properties, boundary conditions, and the like.
S120: and (3) statistically analyzing typical ships in the bridge site water area, and dividing the ships into different tonnage levels according to different types.
S130: building a finite element model of the bridge, performing impact analysis on different parts of the bridge on ships with different tonnages and navigational speeds, and outputting the corresponding probability of collapse of the bridge。
S140: inverting the bridge obtainedProbability of collapseAnd drawing a bridge collapse probability table according to the tonnage, the navigational speed and the impact position of the ship.
It should be noted that the bridge belongs to a large structure, and the influence of the structures corresponding to different positions of the bridge on the overall stability of the bridge is different. Meanwhile, the tonnage of the ship determines the impact strength and the impact position height of the ship; since vessels of similar tonnage are almost the same in terms of structural size, in order to simplify the calculation process and calculation amount of the system, the vessels may be divided into different tonnage levels according to the typical vessels in the bridge site water area, and the structural sizes of the vessels in the respective tonnage levels may be regarded as the same. Therefore, the data volume of the bridge collapse probability table can be simplified, and the active early warning system can acquire the corresponding tonnage level only when the structural size of the ship is detected, so that the probability of bridge collapse corresponding to the tonnage ship is rapidly queried in the bridge collapse probability table.
Based on the above, the bridge collapse probability table records the bridge collapse probability of ships with different tonnages under different impact forces generated at different positions of the bridge. Drawing a bridge collapse probability table theoretically requires impact tests with different forces at different positions of the bridge by using ships with different tonnages; the impact force is influenced by the navigational speed and tonnage of the ship, and under the condition of determining the tonnage of the ship, different impact forces can be obtained through different navigational speeds; and drawing a required bridge collapse probability table according to the data recorded by the test. Although the impact test can acquire data more intuitively, the 1:1 prototype impact test is too high in cost to be implemented, and the reduced-scale test can generate a certain error, so that the test process is simulated by adopting a finite element analysis method. Through finite element analysis, the impacts at different positions of the bridge can be exhausted, and the obtained bridge collapse probability table has higher precision.
Common finite element analysis software includes ANSYS, MIDAS and the like; establishing a finite element model of the bridge in finite element software through the acquired bridge structure parameters; then different impacts are applied to different positions of the model to simulate ships with different tonnages to impact with different forces; and finally, analyzing the impact of the model, judging or calculating the collapse probability of the bridge based on the deformation condition of the model caused by the impact, and recording, so that a required bridge collapse probability table can be obtained. Correspondingly, for inquiring the collapse probability of the bridge, the accurate inquiry can be performed based on the impact position of the ship predicted track and the detected tonnage and navigational speed of the ship, and the impact force of the ship can be calculated through the tonnage and navigational speed of the ship.
It will be appreciated that in practice, the bridge impact will in most cases be that of a single vessel impact, and the bridge collapse probability table described above is for a single vessel impact. Of course, in rare cases, a situation in which multiple vessels strike the bridge may occur; there are mainly two kinds of situations for multiple vessels to strike the bridge.
The method comprises the following steps: risk of multiple vessels striking the bridge at the same time; the method can be regarded as the condition that a plurality of single ships strike simultaneously, and the required collapse probability can be obtained by multiplying the corresponding reduction coefficient after accumulating the collapse probability of the bridge corresponding to each single ship strike.
And two,: and the risk that multiple ships strike the bridge in sequence is that the bridge is not repaired and is subjected to new impact after being impacted. The method can be regarded as the condition of multiple impacts of a single ship, and after each impact, the corresponding bridge damage coefficient can be obtained according to the tonnage and the position of the currently impacted ship; the required collapse rate can be obtained by multiplying the bridge collapse probability corresponding to the last single ship collision by the bridge collapse coefficient corresponding to each collision.
It should be noted that the reduction coefficient and the bridge damage coefficient can be obtained by finite element analysis; wherein, the value of the reduction coefficient is smaller than 1, and the bridge reduction coefficient is larger than 1.
In the present embodiment, in the step S200, the track prediction method of the ship is various, and in the present embodiment, a bidirectional long-short term memory network (BiLSTM) and bayesian optimization (Bayesian Optimization, BO) are preferably adopted) The combined trajectory prediction algorithm (BO-BiLSTM) predicts the trajectory of the vessel. After the corresponding ship prediction model is obtained, the predicted track of the ship in the future 10-30 minutes can be output to be used for calculating the probability of the ship striking the bridge。
Specifically, the ship track prediction comprises the following steps:
s210: and inputting a track sample of the ship to establish and train a BiLSTM track prediction model.
S220: and taking the average value of the root mean square error of the test sample as an objective function of the BiLSTM track prediction model.
S230: and performing super-parameter adjustment of the BiLSTM track prediction model by adopting Bayesian optimization.
S240: and outputting the optimized BO-BiLSTM model by taking the super parameter with the minimum objective function value as the final parameter, and outputting the predicted track of the ship within 10-30 minutes based on the BO-BiLSTM model.
It should be noted that, for the output of the predicted trajectory, if the predicted future time is long, the prediction accuracy of the predicted trajectory for the occurrence of the bridge collision is low; if the predicted future time is shorter, the ship is difficult to adjust the route when the bridge collision possibly occurs; therefore, in the case of combining the above, the future time of outputting the predicted track by the BO-BiLSTM model is 10-30 minutes, preferably 15 minutes.
It should also be noted that long-short term memory network (LSTM) is a deep learning model that effectively solves the problem of gradient disappearance occurring in RNN model training through an internal gate structure based on a Recurrent Neural Network (RNN). In a ship navigation track prediction task, the front and rear moment data mutation of the navigation track is large, the LSTM model cannot acquire information from the rear to the front for training, the data utilization rate is low, and the inherent characteristics of the data cannot be fully utilized.
In the present embodiment, a bidirectional long-short-term memory network (BiLSTM) is introduced for the navigation track prediction model establishment of the target ship. When the result of the track prediction model is output, the bidirectional long-short-term memory network (BiLSTM) can realize the fusion output of the unit output results of the model forward and backward at the current moment, so that the model precision can be effectively improved.
Specifically, as shown in fig. 2, the BiLSTM model is divided into two independent LSTM models, the input sequences are respectively input into the two LSTM models in a positive sequence and a reverse sequence for feature extraction, that is, the forward and backward LSTM models respectively learn the data input in two directions, so as to obtain two output vectors with opposite directions of the hidden layer state; and finally, fusing and splicing the two output vectors (namely the extracted feature vectors) through a weight matrix to obtain final output data serving as a final feature expression of the model. The BiLSTM model can enable feature data obtained at a certain moment to have information between the past and the future, and the efficiency and the performance of data feature extraction are obviously superior to those of the LSTM model.
Meanwhile, the quantity of hidden layer neurons plays a decisive role in the fitting capacity of the model, and the prediction precision and the calculation efficiency of the LSTM model are directly controlled by the network structure super-parameters. Therefore, the prediction performance of the model established by different super parameters is very different, and how to select the proper parameters is very important to the establishment of the model. Currently, the hyper-parameters of the LSTM model are typically selected based on the experience of the researcher and the results of multiple experiments; the randomness is larger, and the prediction performance of the model is affected to a certain extent.
Specifically, the steps of using bayesian optimization of the BiLSTM trajectory prediction model are as follows:
s231: based on the objective function of step S220, the super-parameters to be optimized and the corresponding parameter variation ranges are given; the super parameters to be optimized include the number of hidden layers, the number of training batches and the learning rate.
S232: and optimizing the three super parameters by adopting a Bayesian algorithm.
S234: and taking the super parameter corresponding to the minimum objective function value as a final parameter.
It should be appreciated that by Bayesian optimization of the superparameters, uncertainty of the BiLSTM trajectory prediction model and expectations of the objective function may be minimized. Furthermore, a two-way long-short-term memory network and Bayes optimization combined algorithm (BO-BiLSTM) can find a global optimal solution through fewer sampling points, and the positions and the number of the sampling points can be adaptively adjusted, so that the prediction precision and the calculation efficiency of the track prediction model are greatly improved.
In this embodiment, in step S210, the ship position detected by the radar system on the bridge and the position transmitted by the AIS of the ship are combined to be used as the track sample of the ship. If there are more vessels in the area near the bridge site, the proximity of adjacent vessels may cause the radar system to treat both vessels as one vessel for position detection; the position transmission is required to be carried out by means of the AIS of the ship so as to improve the detection accuracy of the radar system; meanwhile, some small ships may not be provided with AIS, so that the ship cannot be solely supported by AIS, and the radar system and the AIS are combined to judge the position of the ship.
In the present embodiment, in the step S200, the probability of the ship striking the bridge is calculatedThe calculation of (1) comprises the following steps:
s250: different waters are defined according to the characteristics of the bridge area waters channel and the distance from the bridge axis.
S260: acquiring probability of ship striking bridge pier based on predicted track of ship。
S270: selecting corresponding water area risk coefficient according to current water area position of shipProbability of ship striking bridge>。
It will be appreciated that the further the vessel is from the bridge, the easier it is for the vessel to take care of the risk of bumping into the bridge. That is, the farther the ship is from the bridge, the easier the route is to be adjusted, and the lower the probability that the ship actually collides with the bridge. Therefore, when the probability of the ship striking the bridge is calculated, the distance from the position of the water area where the ship is located to the bridge needs to be combined; in the calculation formula, different water area positions correspond to different water area risk coefficients, and the obtained probability that the ship hits the bridge pierMultiplying the risk coefficient of the water area at the corresponding position>The required probability of the ship striking the bridge can be obtained>。
It should also be noted that when predicting the track of a ship, the obtained predicted track is not 100% correct, and the algorithm generally gives the accuracy or probability of the predicted track; wherein the accuracy or possible probability of the predicted track negotiating with the bridge is the probability of the ship striking the pier。
In this embodiment, in order to quantify the calculation result, in step S250, the water areas on both sides of the bridge axis are divided into multiple-level risk water areas from far to near; taking a five-level risk water area as an example, the risk water areas V, IV, III, II and I are respectively. Then in step S270, for the water risk factorThe value of (2) can be obtained according to statistics of historical data or through simulation experiments. As derived from empirical data>The values of (2) are as follows: corresponding>A value of 0.2, corresponding to risk waters IV +.>A value of 0.4, corresponding to risk waters III +.>A value of 0.6, corresponding to risk waters II +.>The value is 0.8, corresponding to risk area I>The value is 1.
In this embodiment, in performing step S400, the calculated probability of risk of ship collision may be based onAnd if the risk grades are classified, the active early warning system of the bridge can send early warning signals of corresponding grades according to different risk grades and the current position of the ship in the risk water area. The risk level is divided into: very low, medium, high and very high; ship collision risk probability corresponding to different risk grades>The value of (2) can be defined according to actual needs; in this embodiment, the risk probability of ship collision corresponding to the risk class is +.>The values of (a) are respectively [0,0.01]]、（0.01，0.05]、（0.05，0. 1]、（0.1，0.5](0.5, 1)]。
In this embodiment, the level of the early warning signal generally includes: no early warning, three-stage early warning, two-stage early warning and one-stage early warning. The no early warning is that the active early warning system does not trigger any early warning, and the three-level early warning is that the active early warning system sends a message notification to the ship; the second-level early warning continuously sends out a flash warning lamp and a VHF call to the ship on the basis of the third-level early warning; the primary early warning sends early warning sound to the ship through the directional loudspeaker on the basis of the secondary early warning.
It should be appreciated that in general, the flashing lights and directional speakers can only be seen and heard when the vessel is relatively close to the bridge; therefore, the secondary early warning and the primary early warning generally occur when the ship is relatively close to the bridge.
Specifically, as shown in fig. 3, when the ship is located in the risk water area v, the level of the early warning signal is no early warning; the active early warning system can not send early warning signals to the ship no matter what risk level the ship is at. When the ship is located in the risk water area IV, the three-level early warning can be triggered only when the ship is in an extremely high risk level, and the active early warning system does not send early warning signals when the ship is in other risk levels. When the ship is located in the risk water area III, the second-level early warning is triggered when the ship exceeds the medium risk level, and the rest risk levels are not early warned. When the ship is located in the risk water area II, the ship is in the medium risk level, the three-level early warning is triggered, the first-level early warning is triggered when the ship exceeds the medium risk level, and all other risk levels are not early warned. When the ship is located in the risk water area I, no early warning is generated when the ship is in the extremely low risk level, the third early warning is triggered when the ship is in the low risk level, the second early warning is triggered when the ship is in the medium risk level, and the first early warning is triggered when the ship is in the medium risk level.
The foregoing has outlined the basic principles, main features and advantages of the present application. It will be appreciated by persons skilled in the art that the present application is not limited to the embodiments described above, and that the embodiments and descriptions described herein are merely illustrative of the principles of the present application, and that various changes and modifications may be made therein without departing from the spirit and scope of the application, which is defined by the appended claims. The scope of protection of the present application is defined by the appended claims and equivalents thereof.
Claims (6)
1. A cross-sea bridge active early warning method based on ship collision risk probability identification is characterized by comprising the following steps:
s100: acquiring a bridge collapse probability table under the collision of the ship bridge;
s200: predicting the track of the ship, and calculating the probability of the ship striking the bridge according to the predicted trackA striking position;
s300: inquiring a bridge collapse probability table based on the tonnage, the navigational speed and the predicted impact position of the ship to obtain the bridge collapse probability corresponding to the impact positionThe method comprises the steps of carrying out a first treatment on the surface of the The risk probability of ship collision->；
S400: sending corresponding early warning signals according to the calculated ship collision risk probability beta and the distance between the ship and the bridge;
in step S100, the acquisition of the bridge collapse probability table includes the following processes:
s110: obtaining structural parameters of a target bridge;
s120: the method comprises the steps of statistically analyzing typical ships in the bridge site water area, and dividing the ships into different tonnage levels according to different types;
s130: building a finite element model of the bridge, performing impact analysis on different parts of the bridge on ships with different tonnages and navigational speeds, and outputting the corresponding probability of collapse of the bridge；
S140: probability of collapse of the bridge to be obtainedDrawing a bridge collapse probability table according to the tonnage, the navigational speed and the impact position of the ship;
in step S200, a track prediction algorithm combining a two-way long-short-term memory network and Bayesian optimization is adopted to predict the track of the ship, and the predicted track of the ship within 10-30 minutes is output for calculating the ship collisionProbability of striking bridge；
The specific ship track prediction comprises the following steps:
s210: inputting a track sample of a ship, and building and training a BiLSTM track prediction model;
s220: taking the average value of the root mean square error of the test sample as an objective function of the BiLSTM track prediction model;
s230: performing super-parameter adjustment of the BiLSTM track prediction model by adopting Bayesian optimization;
s240: outputting an optimized BO-BiLSTM model by taking the super parameter with the minimum objective function value as a final parameter, and outputting a predicted track of the ship within 10-30 minutes in the future based on the BO-BiLSTM model;
in step S200, for the probability of the ship striking the bridgeThe calculation of (1) comprises the following steps:
s250: different waters are defined according to the characteristics of the bridge area water channel and the distance from the bridge axis;
s260: acquiring probability of ship striking bridge pier based on predicted track of ship；
S270: selecting a corresponding water area risk coefficient alpha according to the current water area position of the ship, and enabling the ship to strike the bridge。
2. The cross-sea bridge active early warning method based on ship collision risk probability recognition according to claim 1, wherein the method comprises the following steps: in step S110, the structural parameters of the target bridge include geometric parameters, material properties, and boundary conditions.
3. The cross-sea bridge active early warning method based on ship collision risk probability recognition according to claim 1, wherein the method comprises the following steps: in step S210, the ship position detected by the radar system on the bridge and the position transmitted by the ship' S own AIS are combined as a track sample of the ship.
4. The cross-sea bridge active early warning method based on ship collision risk probability recognition according to claim 1, wherein the method comprises the following steps: in step S250, dividing the water areas on two sides of the axle axis into five-level risk water areas V, IV, III, II and I from far to near;
in step S270, for the value of the water risk coefficient α: the alpha value corresponding to the risk water area V is 0.2, the alpha value corresponding to the risk water area IV is 0.4, the alpha value corresponding to the risk water area III is 0.6, the alpha value corresponding to the risk water area II is 0.8, and the alpha value corresponding to the risk water area I is 1.
5. The cross-sea bridge active early warning method based on ship collision risk probability recognition according to claim 4, wherein the method comprises the following steps: in step S400, risk classification is performed based on the calculated ship collision risk probability β, and then early warning signals of corresponding levels are sent according to different risk levels and the current risk water area position of the ship;
the risk level is divided into: the values of the respective ship collision risk probabilities β are [0,0.01], (0.01, 0.05], (0.05, 0.1 ], (0.1, 0.5], and (0.5, 1], very low, medium, high, and very high.
6. The cross-sea bridge active early warning method based on the ship collision risk probability recognition according to claim 5, wherein the method comprises the following steps: the level of the early warning signal comprises no early warning, three-level early warning, two-level early warning and one-level early warning;
wherein, no early warning is that any early warning is not triggered, three-level early warning is that a message notification is sent to the ship, two-level early warning is that a flash warning lamp, a VHF call and a message notification are sent to the ship, the primary early warning is to send directional loudspeaker, explosion flash warning lamp, VHF call and message notice to the ship;
when the ship is located in the risk water area V, the early warning signals are of no early warning level;
when the ship is located in the risk water area IV, the three-level early warning is triggered only when the ship is at an extremely high risk level;
when the ship is located in the risk water area III, triggering a secondary early warning when the ship exceeds a medium risk level, wherein the rest risk levels are not early warned;
when the ship is located in a risk water area II, triggering three-level early warning when the ship is in a medium risk level, triggering one-level early warning when the ship exceeds the medium risk level, and enabling the rest risk levels to be free of early warning;
when the ship is located in the risk water area I, no early warning is generated when the ship is in the extremely low risk level, the third early warning is triggered when the ship is in the low risk level, the second early warning is triggered when the ship is in the medium risk level, and the first early warning is triggered when the ship is in the medium risk level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410060780.2A CN117576951B (en) | 2024-01-16 | 2024-01-16 | Cross-sea bridge active early warning method based on ship collision risk probability identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410060780.2A CN117576951B (en) | 2024-01-16 | 2024-01-16 | Cross-sea bridge active early warning method based on ship collision risk probability identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117576951A CN117576951A (en) | 2024-02-20 |
| CN117576951B true CN117576951B (en) | 2024-04-16 |
Family
ID=89886678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410060780.2A Active CN117576951B (en) | 2024-01-16 | 2024-01-16 | Cross-sea bridge active early warning method based on ship collision risk probability identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117576951B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101410956B1 (en) * | 2013-02-20 | 2014-06-25 | 이주환 | Implementation Sytem and Method for Collision Prediction, Alarm Methods Between Ships Through Comparison of Integrated Heterogeneous Network |
| CN109949616A (en) * | 2019-03-25 | 2019-06-28 | 同济大学 | A kind of anti-ship of bridge active hits monitoring and warning system |
| KR102112000B1 (en) * | 2019-11-08 | 2020-05-18 | 동명대학교산학협력단 | System for Warning Collision with Marine Bridge and Seaside Facilities based on Radar |
| KR102231343B1 (en) * | 2020-04-24 | 2021-03-24 | (주)아이지오 | Marine warning system for the protection of bridge facilities |
| CN113436468A (en) * | 2021-06-24 | 2021-09-24 | 上海海事大学 | Construction method of bridge collision avoidance system cooperative configuration based on synergetics |
| CN115346399A (en) * | 2022-07-23 | 2022-11-15 | 交通运输部规划研究院 | Bridge ship collision prevention early warning system based on phased array radar, AIS and LSTM network |
| CN115457807A (en) * | 2022-10-25 | 2022-12-09 | 安徽慧软智能科技有限公司 | Ship collision avoidance early warning system based on navigation radar |
| CN115456384A (en) * | 2022-09-01 | 2022-12-09 | 华能国际电力江苏能源开发有限公司 | Ship collision risk grade determining method and equipment |
| CN117392880A (en) * | 2023-09-07 | 2024-01-12 | 广东和立土木工程有限公司 | System and method for evaluating safety risk of ship bridge collision |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11681294B2 (en) * | 2018-12-12 | 2023-06-20 | Here Global B.V. | Method and system for prediction of roadwork zone |
-
2024
- 2024-01-16 CN CN202410060780.2A patent/CN117576951B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101410956B1 (en) * | 2013-02-20 | 2014-06-25 | 이주환 | Implementation Sytem and Method for Collision Prediction, Alarm Methods Between Ships Through Comparison of Integrated Heterogeneous Network |
| CN109949616A (en) * | 2019-03-25 | 2019-06-28 | 同济大学 | A kind of anti-ship of bridge active hits monitoring and warning system |
| KR102112000B1 (en) * | 2019-11-08 | 2020-05-18 | 동명대학교산학협력단 | System for Warning Collision with Marine Bridge and Seaside Facilities based on Radar |
| KR102231343B1 (en) * | 2020-04-24 | 2021-03-24 | (주)아이지오 | Marine warning system for the protection of bridge facilities |
| CN113436468A (en) * | 2021-06-24 | 2021-09-24 | 上海海事大学 | Construction method of bridge collision avoidance system cooperative configuration based on synergetics |
| CN115346399A (en) * | 2022-07-23 | 2022-11-15 | 交通运输部规划研究院 | Bridge ship collision prevention early warning system based on phased array radar, AIS and LSTM network |
| CN115456384A (en) * | 2022-09-01 | 2022-12-09 | 华能国际电力江苏能源开发有限公司 | Ship collision risk grade determining method and equipment |
| CN115457807A (en) * | 2022-10-25 | 2022-12-09 | 安徽慧软智能科技有限公司 | Ship collision avoidance early warning system based on navigation radar |
| CN117392880A (en) * | 2023-09-07 | 2024-01-12 | 广东和立土木工程有限公司 | System and method for evaluating safety risk of ship bridge collision |
Non-Patent Citations (2)
| Title |
|---|
| 内河航道船舶避碰轨迹规划与预测;王垒;宋庭新;;湖北工业大学学报;20190415(02);全文 * |
| 船撞桥最小二乘支持向量机预测方法;罗伟林;邹早建;;交通运输工程学报;20070815(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117576951A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gao et al. | Situational assessment for intelligent vehicles based on stochastic model and Gaussian distributions in typical traffic scenarios | |
| CN109117987B (en) | Personalized traffic accident risk prediction recommendation method based on deep learning | |
| JP6045846B2 (en) | Traffic accident occurrence prediction device, method and program | |
| CN106205156B (en) | A kind of intersection self-healing combination control method for the mutation of part lane flow | |
| CN103557884A (en) | Multi-sensor data fusion early warning method for monitoring electric transmission line tower | |
| CN110570672B (en) | Regional traffic signal lamp control method based on graph neural network | |
| Škorput et al. | Real-time detection of road traffic incidents | |
| CN107985189A (en) | Towards driver's lane change Deep Early Warning method under scorch environment | |
| CN115662113B (en) | Signal intersection man-vehicle game conflict risk assessment and early warning method | |
| Zyryanov et al. | Simulation of evacuation route choice | |
| Elahi et al. | Computer vision based road traffic accident and anomaly detection in the context of Bangladesh | |
| Zou et al. | Short-term traffic flow prediction based on PCC-BiLSTM | |
| CN117576951B (en) | Cross-sea bridge active early warning method based on ship collision risk probability identification | |
| CN114446046A (en) | LSTM model-based weak traffic participant track prediction method | |
| Lv et al. | Research on accident prediction of intersection and identification method of prominent accident form based on back propagation neural network | |
| Ding et al. | Markov chain-based platoon recognition model in mixed traffic with human-driven and connected and autonomous vehicles | |
| Dinh et al. | Real-time queue-end detection on freeways with floating car data: Practice-ready algorithm | |
| Xiao et al. | AIS data analysis for realistic ship traffic simulation model | |
| Tomar et al. | Lane change trajectory prediction using artificial neural network | |
| CN115805865A (en) | Heavy-duty car blind area early warning method, device, equipment and storage medium | |
| CN109191836A (en) | A kind of real-time divided lane vehicle delay prediction technique based on IQA | |
| CN110610611B (en) | Driving safety evaluation method for intelligent network-connected vehicle in mixed-driving traffic flow | |
| Liang | Feature extraction of broken glass cracks in road traffic accident site based on deep learning | |
| He et al. | Model of working ship crossing channel | |
| TWI675347B (en) | Traffic congestion prediction system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 21-1, Building 028, Building 5, No. 15, Lane 587, Juxian Road, High tech Zone, Yinzhou District, Ningbo City, Zhejiang Province, 315000 Patentee after: Ningbo Langda Technology Co.,Ltd. Country or region after: China Address before: 21-1, Building 028, Building 5, No. 15, Lane 587, Juxian Road, High tech Zone, Yinzhou District, Ningbo City, Zhejiang Province, 315000 Patentee before: Ningbo Landa Engineering Technology Co.,Ltd. Country or region before: China |
|
| CP03 | Change of name, title or address |