Marine oil spill emergency treatment system and treatment method thereof

Abstract

The invention discloses a marine oil spill emergency treatment system and a treatment method thereof, and belongs to the technical field of marine environment protection. According to the invention, the unmanned aerial vehicle carries a dual-spectrum camera to carry out cruise scanning to obtain a multispectral image of an oil spill area, an HSV color space segmentation algorithm is utilized to generate an oil spill thermodynamic diagram, an improved Dijkstra algorithm is adopted to plan an optimal recovery path in combination with ship positioning data, instruction synchronization of the unmanned aerial vehicle and a ship is realized through a LoRa communication link, and the recovery efficiency of the oil spill area is improved. And the rotating speed of the oil suction disc is dynamically adjusted based on data of the infrared oil film thickness sensor. According to the emergency treatment system, an air-sea integrated monitoring network is constructed, a hierarchical decision mechanism and a self-adaptive control strategy are established, the problems that monitoring and execution are separated, equipment adjustment is lagged and the cooperation efficiency is low in traditional oil spill treatment are solved, monitoring-decision-execution closed-loop control is achieved, the oil spill recovery response time is shortened to be within 10 minutes, and the oil spill recovery efficiency is improved. The recovery rate of the thin oil layer is increased to 72% or above, and the communication packet loss rate is reduced to 3% or below.

Description

Technical Field

[0001] The present invention relates to the technical field of marine environmental protection, and more particularly to a marine oil spill emergency treatment system and a treatment method thereof. Background Art

[0002] The current marine oil spill treatment has the following technical defects:

[0003] Disconnect between monitoring and execution: Traditional satellite monitoring requires over two hours to transmit data, preventing vessels on the scene from obtaining real-time data on changes in the spill boundary, leading to significant response delays. For example, in one offshore oil spill, satellite data transmission took two hours and 15 minutes, missing the optimal recovery opportunity and causing the spill's spread to triple.

[0004] Equipment adjustment lag: The skimmer speed relies on manual observation and adjustment, resulting in a recovery rate of less than 40% for thin oil layers. For example, in a certain marine oil spill, the recovery rate of a manually adjusted skimmer was only 38% when the oil film thickness was 1.5mm.

[0005] Inefficient collaboration: UAVs and ships use one-way communication, resulting in command transmission delays exceeding 30 seconds and packet loss rates as high as 15%. For example, in a maritime oil spill incident, one-way communication caused a 45-second delay in ship routing and an oil slick tracking error exceeding 50 meters. Summary of the Invention

[0006] 1. Technical problem to be solved by the invention

[0007] In response to the above-mentioned problems existing in the existing marine oil spill treatment technology, the present invention provides a marine oil spill emergency treatment system and treatment method. Through integrated air-sea monitoring, hierarchical decision-making and adaptive control, a complete oil spill treatment closed-loop system is constructed, providing an efficient and reliable technical solution for marine oil spill emergency treatment.

[0008] 2. Technical solution

[0009] In order to achieve the above object, the technical solution provided by the present invention is:

[0010] A marine oil spill emergency treatment system and treatment method thereof, comprising the following steps:

[0011] (1) Use the camera carried by the drone to cruise and scan the oil spill area to obtain multispectral image data;

[0012] (2) The image data is processed using the HSV color space segmentation algorithm to generate an oil spill heat map:

[0013] (3) generating an optimal recovery path for the ship by using an improved Dijkstra algorithm based on the positioning coordinates of the ship's GPS or Beidou satellite navigation system and the coordinates of the characteristic points of the heat map;

[0014] (4) Establishing a LoRa communication link between the UAV and the ship, transmitting the path coordinate point sequence and equipment control instructions;

[0015] (5) Dynamically adjust the speed of the ship's oil suction plate based on the oil film thickness data fed back by the infrared oil film thickness sensor.

[0016] In a further processing method, the camera in step (1) is a visible light / infrared dual-spectrum camera;

[0017] The HSV color space segmentation algorithm in step (2) is as follows: converting the RGB image into the HSV space, setting the oil spill feature thresholds H∈[20,40], S∈[0.2,0.6], and V∈[0.3,0.8], and performing morphological closing operations on the pixels that meet the thresholds to generate a continuous oil film area;

[0018] The cost function of the improved Dijkstra algorithm in step (3) is F(n)=α·distance(n)+β·oil film concentration(n)+γ·ocean current influence(n), where α+β+γ=1, and n is a path node;

[0019] In the step (5), the dynamic adjustment satisfies: when the oil film thickness δ≤2mm, the oil suction plate rotation speed V=200 rpm; when 2mm<δ≤5mm, V=200+50×(δ-2); when δ>5mm, V=350 rpm.

[0020] In a further processing method, the UAV cruise scanning height in step (1) is 50±1 meters, and a zigzag path cruise is adopted, with a single-side coverage width of 100±2 meters.

[0021] Further processing method, the improved Dijkstra algorithm in step (3) also considers the ship's roll angle θ, and processes the ship's attitude data through the Kalman filter, and the roll compensation coefficient K_comp=1+0.05×|θ|+0.02×δ, and the final speed V - set=V - base×K - comp.

[0022] In a further processing method, the LoRa communication link in step (4) adopts the E32-900T module, the communication cycle is 200ms, and the communication packet loss rate is ≤3%.

[0023] In a further processing method, in step (5), the measuring range of the infrared oil film thickness sensor is 0-10 mm, the measuring accuracy is ±0.5 mm, and the sampling frequency is 1 kHz.

[0024] Further processing method, preferably the steps are:

[0025] (1) The drone cruise scans to obtain multispectral images of the oil spill area, with a flight altitude of 50±1 meters, a zigzag path, a single-side coverage width of 100 meters, and a sampling frequency of 1 Hz.

[0026] (2) The image data is converted into HSV space, and the oil spill feature thresholds are set as H∈[20,40], S∈[0.2,0.6], and V∈[0.3,0.8]. The continuous oil film area thermal map is generated through morphological closing operation.

[0027] (3) The coordinates of the characteristic points of the heat map are extracted and combined with the GPS positioning data of the ship. The cost function F(n) = α·distance(n) + β·oil film concentration(n) + γ·ocean current influence(n) (α = 0.4, β = 0.3, γ = 0.3) is used to plan the optimal recovery path.

[0028] (4) The path point sequence and control instructions are transmitted through the LoRa communication link (E32-900T module), with a communication cycle of 200ms and a packet loss rate of ≤3%.

[0029] (5) The infrared oil film thickness sensor (measuring range 0-10 mm, accuracy ±0.5 mm) provides real-time feedback data. The oil suction plate speed is adjusted according to the rule of V = 200 rpm when δ ≤ 2 mm, V = 200 + 50 × (δ - 2) when 2 mm < δ ≤ 5 mm, and V = 350 rpm when δ > 5 mm. The speed is corrected in combination with the roll compensation coefficient K_comp = 1 + 0.05 × |θ| + 0.02 × δ.

[0030] A marine oil spill emergency treatment system, comprising:

[0031] The drone cruise module is used to carry a dual-spectral camera to obtain image data of the oil spill area and realize multi-spectral data collection;

[0032] Heat map generation module, used to process images based on the HSV color space segmentation algorithm to generate oil spill heat maps;

[0033] Path planning module, used to plan the optimal path based on the ship positioning and the coordinates of the heat map feature points using the improved Dijkstra algorithm;

[0034] Command transmission module, used to establish the LoRa communication link between the drone and the ship;

[0035] The power regulation module is used to dynamically adjust the oil suction plate speed according to the infrared oil film thickness data.

[0036] For a further emergency response system, the ship is equipped with a GPS or Beidou satellite navigation system positioning module and attitude sensor; the positioning module has a positioning accuracy of ±1 meter, and the attitude sensor has a roll angle measurement range of ±15° and an accuracy of ±0.2°, and can obtain position and attitude data in real time.

[0037] The entire system consists of:

[0038] (1) Integrated air and sea monitoring network

[0039] Drones equipped with visible light / infrared dual-spectrum cameras (such as the DJI M300 RTK equipped with the Zenmuse H20T infrared camera) can achieve multispectral data collection.

[0040] The ship is equipped with a GPS positioning module (positioning accuracy ±1 meter) and an attitude sensor (roll angle measurement range ±15°, accuracy ±0.2°) to obtain position and attitude data in real time.

[0041] (2) Hierarchical decision-making mechanism

[0042] The drone executes a fast image segmentation algorithm to achieve real-time identification of oil spill areas based on the HSV color space.

[0043] The ship runs the improved Dijkstra algorithm and plans the optimal path based on distance, oil film concentration and ocean current influence.

[0044] (3) Adaptive control strategy

[0045] The oil suction plate speed is dynamically adjusted based on the infrared oil film thickness sensor data to adapt to different oil film thicknesses.

[0046] The ship roll compensation strategy is adopted and the Kalman filter algorithm is used to eliminate the influence of ship roll on the recovery efficiency.

[0047] 3. Beneficial effects

[0048] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0049] The marine oil spill emergency response system and method of the present invention significantly shortens response time: the monitoring-decision-execution closed-loop time is compressed from 60 minutes to within 10 minutes. In a simulation experiment, the time from oil spill discovery to vessel recovery was only 8 minutes and 42 seconds.

[0050] Recovery efficiency has been significantly improved: the recovery rate of thin oil layers (δ≤2mm) has increased from 40% to 72%, and the overall oil spill recovery efficiency has increased to over 85%;

[0051] Enhanced communication reliability: The communication packet loss rate has been reduced from the industry average of 15% to below 3%, and the communication cycle is 200ms, ensuring real-time transmission of instructions.

[0052] Improved environmental adaptability: Through the roll compensation strategy, the speed adjustment error is ≤5% when the ship's roll angle is ±10°, ensuring recovery stability in complex sea conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of the operation of the marine oil spill emergency treatment system of a specific embodiment;

[0054] Figure 2 This is a comparison chart of HSV color space segmentation effects;

[0055] Figure 3 Schematic diagram of ship path planning.

[0056] In the figure: 101 - UAV cruise module; 102 - thermal map generation module; 103 - path planning module; 104 - command transmission module; 105 - power regulation module; 106 - ship oil suction pan. DETAILED DESCRIPTION

[0057] In order to further understand the content of the present invention, the invention is described in detail with reference to the accompanying drawings.

[0058] Example 1

[0059] The marine oil spill emergency treatment system and treatment method of this embodiment are applied to the treatment of thin oil layer oil spills in nearshore waters:

[0060] like Figure 1 As shown, the emergency handling system includes a UAV cruise module 101, a heat map generation module 102, a path planning module 103, a command transmission module 104 and a power regulation module 105;

[0061] Drone Cruise Module 101: DJI M300 RTK, equipped with Zenmuse H20T infrared camera, flight altitude 50 meters, cruise speed 10m / s, such as Figure 3 As shown, it shows the broken line path of the ship from point A to point B (the center of the oil film), zigzag scanning, and a single-side coverage width of 100 meters.

[0062] Vessel: 5-meter-long catamaran workboat equipped with SK-5 rotary ship oil suction cup 106 (rated power 5.5kW, rated speed 300rpm), GPS or Beidou satellite navigation system positioning module (positioning accuracy ±1 meter), attitude sensor (roll angle measurement range ±15°, accuracy ±0.2°), infrared oil film thickness sensor (measuring range 0-10mm, accuracy ±0.5mm, sampling frequency 1kHz).

[0063] Communication module: LoRa E32-900T, communication distance 2km, communication cycle 200ms.

[0064] Workflow:

[0065] Take-off phase: The drone takes off from the shore platform, rises to an altitude of 50 meters within 30 seconds, and enters cruise mode.

[0066] Monitoring phase: 1km near the shore in a zigzag path 2 The water area is scanned, and a multispectral image is obtained every 10 seconds and transmitted to the ground station.

[0067] Data processing: After the ground station receives the RGB image, it executes the HSV segmentation algorithm:

[0068] Python#HSV segmentation pseudocode

[0069] hsv=cv2.cvtColor(rgb_img,cv2.COLOR_RGB2HSV)

[0070] mask=cv2.inRange(hsv,(20,50,50),(40,255,255))

[0071] kernel=np.ones((5,5),np.uint8)

[0072] closed_mask=cv2.morphologyEx(mask,cv2.MORPH_CLOSE,kernel).

[0073] like Figure 2 As shown, the left is the original RGB image, and the right is the binary thermal map after HSV segmentation. The thermal map generation module 102 generates a binary thermal map, extracts the coordinates of the oil film contour vertices, and calculates the oil film area of ​​about 0.2 km 2 The average thickness is 1.8mm. The working principle is: the drone detects the pollution on the sea surface and identifies Figure 2 The picture will be separated into black and white by HSV technology, and then these black dots will be written in the code of the above program. When it is run, it will identify the illustrated area, making it easier to accurately treat the pollution in this sea area later.

[0074] The path planning module 103 plans the path generation: Figure 3The figure shows the optimal path planned for a ship from point A to point B (the center of the oil slick), taking into account the influence of oil slick concentration and ocean currents. The ship's current position is A (120.5°E, 31.3°N), the oil slick's center is B (120.52°E, 31.31°N), and the ocean current is 0.5 m / s northeast. Using the modified Dijkstra algorithm with a cost function of α = 0.4, β = 0.3, and γ = 0.3, a broken-line path A → C → B → D is planned, with a total distance of 1.2 km and an estimated travel time of 8 minutes.

[0075] The instruction transmission module 104 performs instruction transmission: the path point sequence (coordinates of A, C, B, and D) and the initial control instruction (oil suction plate speed 200 rpm) are sent through the LoRa module. The communication delay is 180 ms and there is no packet loss.

[0076] Combine Figure 3 The power regulation module 105 performs dynamic adjustment: when the ship reaches point C, the infrared sensor detects the oil film thickness δ = 1.8 mm, and the speed is maintained at 200 rpm; when it reaches point D, δ = 2.3 mm, and the speed is calculated as V = 200 + 50 × (2.3 - 2) = 215 rpm. At the same time, the ship's roll angle θ = 3°, the compensation coefficient K_comp = 1 + 0.05 × 3 + 0.02 × 2.3 = 1.196, and the final speed V_set = 215 × 1.196 ≈ 257 rpm.

[0077] Verification results:

[0078] The time from the discovery of the oil spill to the start of vessel recovery took 8 minutes and 42 seconds, which is 86% shorter than traditional methods.

[0079] The oil film thickness recognition error is ±0.3mm, and the recognition accuracy is 98.5%.

[0080] The communication cycle is 200ms, the packet loss rate is 2.3%, and the command transmission is stable.

[0081] The recovery operation lasted 30 minutes, and the recovery rate of thin oil layers (δ≤2mm) was 72.3%, an increase of 80.7% compared with manual adjustment.

[0082] Example 2

[0083] The marine oil spill emergency treatment system and treatment method of this embodiment have the same basic equipment configuration and steps as Example 1, with the difference or improvement being that they are applied to the treatment of medium oil film thickness oil spills (Bohai Bay case):

[0084] Scene setting:

[0085] Oil spill area: simulated near a port in Bohai Bay, with an oil film area of ​​0.5km 2, average thickness 3.5mm, current speed 0.8m / s, direction southeast.

[0086] The equipment configuration is the same as that of Example 1, with the addition of a current sensor (measuring range 0-2 m / s, accuracy ±0.1 m / s).

[0087] Key steps:

[0088] HSV segmentation: After the original image is transformed into HSV and morphologically processed, the oil film area recognition accuracy is 97.8%, and the contour extraction error is ≤5 meters.

[0089] Path planning: In the cost function, γ = 0.4 (the weight of the influence of ocean currents is increased). The ship travels from position P (118.9°E, 38.7°N) to the center of the oil film Q (118.92°E, 38.71°N). The planned path takes into account the push of ocean currents and adopts an S-shaped route. The total distance is 1.5 km and the estimated sailing time is 10 minutes.

[0090] Dynamic adjustment: When δ = 3.5 mm, the basic speed V_base = 200 + 50 × (3.5 - 2) = 275 rpm, the ship roll angle θ = 5°, K_comp = 1 + 0.05 × 5 + 0.02 × 3.5 = 1.32, V_set = 275 × 1.32 = 363 rpm. Since the upper limit of the speed is 350 rpm, the limit is 350 rpm.

[0091] Technical Effects

[0092] The response time is 9 minutes and 15 seconds, meeting the requirement of within 10 minutes.

[0093] The oil film recovery efficiency is 85.6%, of which the recovery rate of 3-5mm thick oil film is 82.1%.

[0094] After ship roll compensation, the speed fluctuation is ≤±8rpm, and the recovery stability is good

[0095] Example 3

[0096] The marine oil spill emergency treatment system and treatment method of this embodiment have the same basic equipment configuration and steps as the first embodiment, with the difference or improvement being that they are applied to oil spill treatment in a strong ocean current environment:

[0097] Complex environment parameters:

[0098] Oil spill area: simulates the area near a drilling platform in the South China Sea, with a current speed of 1.2m / s, southwest direction, wave height of 1.5m, and oil film thickness of 5.5mm (exceeding 5mm in some areas).

[0099] Equipment configuration: The UAV has a wind resistance level of ≥6, the ship is equipped with anti-roll fins, and the infrared sensor is equipped with a waterproof shell (IP68); other equipment configurations are the same as in Example 1.

[0100] Assignment focus:

[0101] Path planning: The ocean current influence weight γ = 0.5, the cost function F(n) = 0.3·distance(n) + 0.2·oil film concentration(n) + 0.5·ocean current influence(n), and a detour route is planned to avoid the ship sailing against the strong current. The total distance is 2.1 km and the sailing time is 14 minutes (including the influence of ocean current).

[0102] Dynamic adjustment: When δ>5mm, the basic speed is 350rpm, the ship's roll angle θ=8°, K_comp=1+0.05×8+0.02×5.5=1.51, V_set=350×1.51=528.5rpm. Since the speed limit is 350rpm, the overcurrent protection is triggered, the frequency is reduced by 50% to 175rpm, and the anti-roll fins are activated to reduce the roll angle.

[0103] Communication optimization: Using frequency hopping technology, the communication packet loss rate is reduced from 4.5% to 2.8%, ensuring reliable command transmission.

[0104] Verify data

[0105] It took 9 minutes and 50 seconds from the discovery of the oil spill to the start of recovery, which still met the 10-minute response requirement even in a strong current environment.

[0106] The recovery rate of thick oil layers (δ>5mm) is 88.3%, and the overall recovery efficiency is 85.2%.

[0107] The roll compensation strategy enables the oil suction plate to work effectively for 92% of the time when the ship's wave height is 1.5 meters, an increase of 40% compared to traditional methods.

[0108] Summary of technical effects of Examples 1-3:

[0109] Response time: The average response time in the three embodiments was 8 minutes and 59 seconds, all ≤10 minutes, which is 91.6% shorter than the more than 2 hours of traditional satellite monitoring.

[0110] Recovery efficiency: The average recovery rate of thin oil layers (δ≤2mm) is 72.1%, the average recovery rate of medium oil films (2mm<δ≤5mm) is 82.4%, and the average recovery rate of thick oil layers (δ>5mm) is 88.3%. The overall recovery efficiency is 85.2%, far exceeding the 40-60% of traditional methods.

[0111] Communication performance: The average packet loss rate of the LoRa communication link is 2.5%, the communication cycle is 200ms, and the command transmission delay is ≤200ms, meeting the real-time collaboration requirements.

[0112] Environmental adaptability: In a complex environment with a roll angle of ±10°, a current speed of 1.2m / s, and a wave height of 1.5 meters, the system can still work stably, and the recovery efficiency fluctuation is ≤5%, demonstrating strong environmental adaptability.

[0113] The marine oil spill emergency response system and treatment method of Examples 1-3 construct a complete oil spill response closed-loop system through integrated air-sea monitoring, hierarchical decision-making and adaptive control, effectively solving the defects of traditional technologies and providing an efficient and reliable technical solution for marine oil spill emergency response.

[0114] The above is a schematic description of the present invention and its embodiments, which is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure and manufacturing steps are not limited thereto. Therefore, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without creatively designing them, they shall fall within the scope of protection of the present invention.

Claims

1. A method for treating a marine oil spill emergency treatment system, characterized in that: The following steps are involved: (1) Use the camera onboard the drone to conduct patrol scans of the oil spill area; (2) Using the HSV color space segmentation algorithm to process the image data: (3) generating an optimal recovery path for the ship by using an improved Dijkstra algorithm based on the positioning coordinates of the ship's GPS or Beidou satellite navigation system and the coordinates of the characteristic points of the heat map; (4) Establishing a LoRa communication link between the UAV and the ship, transmitting the path coordinate point sequence and equipment control instructions; (5) Dynamically adjust the speed of the ship's oil suction plate based on the oil film thickness data fed back by the infrared oil film thickness sensor.

2. The processing method according to claim 1, characterized in that: The camera in step (1) is a visible light / infrared dual-spectrum camera; The HSV color space segmentation algorithm in step (2) is as follows: converting the RGB image into the HSV space, setting the oil spill feature thresholds H∈[20,40], S∈[0.2,0.6], and V∈[0.3,0.8], and performing morphological closing operations on the pixels that meet the thresholds to generate a continuous oil film area; The cost function of the improved Dijkstra algorithm in step (3) is F(n)=α·distance(n)+β·oil film concentration(n)+γ·ocean current influence(n), where α+β+γ=1, and n is a path node; In the step (5), the dynamic adjustment satisfies: when the oil film thickness δ≤2mm, the oil suction plate rotation speed V=200 rpm; when 2mm<δ≤5mm, V=200+50×(δ-2); when δ>5mm, V=350 rpm.

3. The processing method according to claim 1, characterized in that: In the step (1), the UAV cruise scan height is 50±1 meters, a zigzag path cruise is adopted, and the single-side coverage width is 100±2 meters.

4. The processing method according to claim 1, characterized in that: The improved Dijkstra algorithm in step (3) also considers the ship's roll angle θ, and processes the ship's attitude data through a Kalman filter, with a roll compensation coefficient K_comp = 1 + 0.05 × |θ| + 0.02 × δ, and a final rotation speed V-set = V-base × K-comp.

5. The processing method according to claim 1, characterized in that: In step (4), the LoRa communication link adopts the E32-900T module, the communication cycle is 200ms, and the communication packet loss rate is ≤3%.

6. The processing method according to claim 1, characterized in that: In step (5), the measurement range of the infrared oil film thickness sensor is 0-10 mm, the measurement accuracy is ±0.5 mm, and the sampling frequency is 1 kHz.

7. A marine oil spill emergency treatment system according to any one of claims 1 to 7, characterized in that: include: The drone cruise module is used to carry a dual-spectral camera to obtain image data of the oil spill area; Heat map generation module, used to process images based on the HSV color space segmentation algorithm to generate oil spill heat maps; Path planning module, used to plan the optimal path based on the ship positioning and the coordinates of the heat map feature points using the improved Dijkstra algorithm; Command transmission module, used to establish the LoRa communication link between the drone and the ship; The power regulation module is used to dynamically adjust the oil suction plate speed according to the infrared oil film thickness data.

8. The marine oil spill emergency treatment system according to claim 7, characterized in that: The vessel is equipped with a GPS or Beidou satellite navigation system positioning module and an attitude sensor; the positioning module has a positioning accuracy of ±1 meter, and the attitude sensor has a roll angle measurement range of ±15° and an accuracy of ±0.2°.

Images