Offshore platform system for treating marine litter

Abstract

According to the offshore platform system for processing the marine litter, the litter processing module is arranged, the unloading task is executed, the ship information of the ship transportation module is collected, the ship information is recognized, the position of the marine litter and the position of an obstacle are determined, and the marine litter is processed according to the position of the marine litter and the position of the obstacle. The marine litter is unloaded to the litter treatment production line unit, the litter treatment production line unit executes a litter treatment task and treats the marine litter, the ship transportation module does not need to frequently return to a shore-based port, and the cleaning efficiency is improved; and the multi-source energy supply module obtains a historical load power sequence at a preset time interval and task parameters of all tasks needing to be executed, predicts required power after the preset time interval according to the historical load power sequence and the task parameters, adjusts output power of various energy sources according to the required power and priorities of the various energy sources, and supplies power to the multi-source energy supply module. Various energy sources are provided for power supply, various energy source output power is intelligently regulated and controlled, and full utilization of energy sources is facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ocean, in particular to a sea platform system for processing marine garbage. BACKGROUND

[0002] Marine garbage can cause ecological damage, fishery loss and channel risk. In recent years, with the development of unmanned technology, the cleaning of marine surface garbage has gradually developed towards the direction of autonomous fishing of unmanned garbage recycling ships. However, the unmanned garbage recycling ship still needs to return to the shore base frequently for charging or garbage unloading, which seriously affects the cleaning efficiency. SUMMARY

[0003] The embodiment of the present application provides a sea platform system for processing marine garbage to solve at least one problem in the related art, and the technical scheme is as follows: The embodiment of the present application provides a sea platform system for processing marine garbage, comprising: A ship transportation module for collecting marine garbage; A garbage disposal module comprising an automatic unloading unit and a garbage disposal production line unit, the automatic unloading module is used to perform an unloading task, collect ship information of the ship transportation module, identify the ship information to determine the position of marine garbage and the position of obstacles, unload the marine garbage to the garbage disposal production line unit according to the position of marine garbage and the position of obstacles, and the garbage disposal production line unit is used to perform a garbage disposal task and process marine garbage; A multi-source energy supply module for obtaining a historical load power sequence of a preset time interval and task parameters of all tasks to be performed, predicting a demand power after a preset time interval according to the historical load power sequence and the task parameters, and adjusting the output power of various energy sources according to the demand power and the priority of various energy sources.

[0004] In an embodiment, the identification of the ship information to determine the position of marine garbage and the position of obstacles, and the unloading of marine garbage to the garbage disposal production line unit according to the position of marine garbage and the position of obstacles comprises: The image of the ship information is identified by a target detection network and a binocular disparity, the three-dimensional coordinate position of marine garbage is determined, and the position of obstacles is identified according to the obstacle data of the ship information; The moving path of the mechanical arm is determined according to a path planning algorithm, the three-dimensional coordinate position and the position of obstacles; The control parameters of the mechanical arm are determined according to the three-dimensional coordinate position and the mechanical parameters of the mechanical arm; According to the control parameter and the moving path, the marine garbage is unloaded by the mechanical arm to the garbage processing production line unit.

[0005] In an embodiment, the control parameter of the mechanical arm is determined according to the three-dimensional coordinate position and the mechanical parameter of the mechanical arm, including: According to the three-dimensional coordinate position, the length of the connecting rod of the shoulder joint and the length of the connecting rod of the elbow joint of the mechanical arm, the angle of the shoulder joint and the elbow joint and the corresponding connecting rod is calculated respectively by using a two-dimensional plane inverse kinematics model; According to the angle and the three-dimensional coordinate position, the contact area, the actual posture and the actual speed of the end effector of the mechanical arm are determined when the marine garbage is clamped by the end effector, and the target clamping force is determined according to the contact area, the preset friction coefficient and the preset garbage compression strength; According to the preset impedance stiffness coefficient, the preset impedance damping coefficient, the preset expected posture, the preset expected speed, the actual posture and the actual speed of the end effector, the control instruction of the end effector is determined by an impedance control model; The control parameter includes the angle, the target clamping force and the control instruction.

[0006] In an embodiment, the marine garbage is processed, including: The marine garbage is compressed by a screw compressor, and in the process of compression, the theoretical pressure distribution is determined by a garbage compression model, and the screw rotation speed is controlled according to the theoretical pressure distribution and a PID control algorithm; The compressed marine garbage is incinerated by a pyrolysis incinerator; The sewage generated in the process of compression and incineration is filtered by an MBR membrane bioreactor.

[0007] In an embodiment, the energy sources include photovoltaic, marine wave energy, battery energy storage, power grid and diesel generator; and the output power of various energy sources is adjusted according to the demand power and the priority of various energy sources, including: According to the power generation cost of various energy sources, the absolute value of the battery charge and discharge power and the corresponding start state value whether the diesel generator is started, a power generation cost function is constructed; Based on the constraint condition and the priority of various energy sources, the output power of various energy sources is adjusted based on column generation method to minimize the function value of the power generation cost function; The priority of the energy sources from high to low is photovoltaic, ocean wave energy, battery energy storage, power grid and diesel generator, and the constraint conditions include that the sum of the final output power of various energy sources is greater than or equal to the demand power, the electric quantity of the total power supply of the multi-source energy supply module is in a preset electric quantity range, and the change rate of the wave power of the ocean wave energy is in a preset change range.

[0008] In an embodiment, the multi-source energy supply module further comprises a wireless charging unit, and the wireless charging unit is configured to: In the process of adjusting the output power of various energy sources for charging the ship transportation module, wireless power transmission is performed through magnetic resonance coupling technology; In the process of power transmission, the resonance frequency is kept within a preset frequency range by dynamically adjusting the resonance capacitance; The transmission efficiency of wireless power transmission is determined, and if the transmission efficiency is less than the efficiency threshold, the current output power of various energy sources is excluded, and the step of adjusting the output power of various energy sources based on the constraint conditions and the priority of various energy sources is returned. The function value of the function value of the power generation cost function is minimized as the target based on the column generation method until the transmission efficiency is greater than or equal to the efficiency threshold.

[0009] In an embodiment, the offshore platform system for processing marine garbage further comprises an intelligent docking module, and the intelligent docking module is configured to: Coarse positioning is performed through ultra-wideband to determine the first coordinate position of at least one ship in the ship transportation module; The speed vector of the ship is obtained through the inertial measurement unit, the first coordinate position and the speed vector are used for pre-adjustment of the hydraulic cushion pile, and the pre-charging pressure value of the hydraulic cushion pile is determined; The point cloud data of the ship is obtained through the laser radar unit, the point cloud data is used for pre-loading of the hydraulic cushion pile, the initial extension amount of the hydraulic cushion pile is determined, and the hydraulic cushion pile is controlled into the buffering preparation state according to the pre-charging pressure value and the initial extension amount; In the process that the ship contacts with the hydraulic cushion pile, the position deviation and the change rate of the position deviation between the ship and the hydraulic cushion pile are determined in real time, and the control parameters of the hydraulic cushion pile are output in real time according to the position deviation, the change rate and the PID algorithm.

[0010] In an embodiment, the pre-loading of the hydraulic cushion pile through the point cloud data to determine the initial extension amount of the hydraulic cushion pile comprises: The rotation parameters and translation parameters of the hydraulic cushion pile are iteratively processed through the point cloud data and a preset template point cloud to determine the minimum registration error; Determine the initial telescopic amount according to the product of the minimum registration error and the arm length of the hydraulic cushion pile.

[0011] In an embodiment, the real-time output of the control parameter of the hydraulic cushion pile according to the position deviation, the change rate and the PID algorithm comprises: Map the position deviation and the change rate to a preset number of fuzzy sets as input variables; Variable fuzzification is performed on the input variables by a preset number of series membership functions, and the control parameter of the hydraulic cushion pile is output in real time according to the preset rules corresponding to the variable-fuzzified input variables and the control parameter.

[0012] In an embodiment, the offshore platform system for processing marine garbage further comprises a central control module, which is configured to: Obtain a state information sequence of the ship transportation module, the garbage processing module, the multi-source energy supply module and the intelligent docking module; Output a fault index according to the state information sequence and an LSTM model with a fusion attention mechanism; When the fault index is greater than or equal to a fault threshold, a fault is prompted and / or a redundant device is switched.

[0013] The beneficial effects of the above technical solutions at least include: The offshore platform system sets a garbage processing module to perform unloading tasks, collects ship information of the ship transportation module, identifies the position of marine garbage and the position of obstacles according to the ship information, and unloads the marine garbage to a garbage processing production line unit according to the position of marine garbage and the position of obstacles. The garbage processing production line unit is configured to perform garbage processing tasks and process marine garbage. The ship transportation module does not need to frequently return to a shore-based port, thereby improving the cleaning efficiency. The multi-source energy supply module obtains a historical load power sequence at a preset time interval and task parameters of all tasks to be performed, predicts the demand power after the preset time interval according to the historical load power sequence and the task parameters, and adjusts the output power of various energy sources according to the demand power and the priority of various energy sources, thereby providing multi-energy power supply and intelligent regulation of various energy output powers, which is conducive to fully and effectively utilizing energy.

[0014] The above summary is only for the purpose of the description and is not intended to limit in any way. In addition to the illustrative aspects, embodiments and features described above, those aspects, embodiments and features will be readily apparent to those skilled in the art by reference to the drawings and the following detailed description, and it is intended that the scope of the application disclosed herein be interpreted by the claims. BRIEF DESCRIPTION OF DRAWINGS

[0015] In the drawings, like reference numerals refer to like elements throughout the various figures. The drawings are not necessarily to scale, and the emphasis is on the functional description of the drawings. It should be understood that the drawings only depict some embodiments in accordance with the disclosure and should not be considered limiting of the scope of the disclosure.

[0016] Figure 1 Block diagram of a marine platform system for processing marine litter according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] In the following, only some exemplary embodiments are described in brief. As the person skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present application. Therefore, the drawings and the description are to be considered as being exemplary in nature and not as being limiting.

[0018] REFERENCE Figure 1 Fig. 1 shows a block diagram of a marine platform system for processing marine litter according to an embodiment of the present application, which can include a vessel transportation module, a litter processing module, a multi-source energy supply module, an intelligent docking module and a central control module on the platform. Wherein: The vessel transportation module is configured to collect marine litter. The litter processing module includes an automated unloading unit and a litter processing production line unit. The automated unloading module is configured to perform unloading tasks, collect vessel information of the vessel transportation module, identify the position of marine litter and the position of obstacles according to the vessel information, unload the marine litter to the litter processing production line unit according to the position of marine litter and the position of obstacles, and the litter processing production line unit is configured to perform litter processing tasks and process marine litter. The multi-source energy supply module is configured to obtain a historical load power sequence of a preset time interval and task parameters of all tasks to be performed, predict a demand power after the preset time interval according to the historical load power sequence and the task parameters, and adjust the output power of various energy sources according to the demand power and the priority of various energy sources.

[0019] The technical scheme of the embodiment of the present application is characterized in that the offshore platform system is provided with a garbage disposal module, performs an unloading task, collects ship information of a ship transportation module, identifies the ship information to determine the position of marine garbage and the position of an obstacle, unloads the marine garbage to a garbage disposal production line unit according to the position of the marine garbage and the position of the obstacle, and the garbage disposal production line unit is used to perform a garbage disposal task and process the marine garbage. The ship transportation module does not need to frequently return to a shore-based port, and the cleaning efficiency is improved. The multi-source energy supply module obtains a historical load power sequence of a preset time interval and task parameters of all tasks to be performed, predicts the demand power after the preset time interval according to the historical load power sequence and the task parameters, adjusts the output power of various energy sources according to the demand power and the priority of various energy sources, and provides multi-energy power supply and intelligent regulation and control of the output power of various energy sources, which is beneficial to fully and effectively utilize energy.

[0020] In the embodiment of the present application, the ship transportation module includes at least one ship, which is used to collect marine garbage and transport the marine garbage to the garbage disposal module on the platform.

[0021] In the embodiment of the present application, the intelligent docking module is used to realize sub-centimeter positioning of the ship through multi-sensor fusion and dynamically absorb the impact energy when the ship approaches the platform. The intelligent docking module includes hydraulic cushion piles (4 groups), binocular vision sensors (4), cable winches (8), laser radar units (4), and inertial measurement units. The hydraulic cushion piles are symmetrically distributed at the edges of the platform, the binocular vision sensors and the laser radars are installed at the four corners of the platform, and the winches are embedded in the base of the platform.

[0022] In the embodiment of the present application, the automatic unloading unit of the garbage disposal module is used to unload the marine garbage on the ship, and the garbage disposal production line unit is used to process the marine garbage. The automatic unloading unit includes a three-degree-of-freedom mechanical arm (maximum load 500 kg), a six-dimensional force sensor (ATI Gamma), a vacuum chuck clamp (diameter 1.2 m), and a laser obstacle avoidance module (SICK LMS511). The base of the mechanical arm is fixed at the entrance of the garbage disposal area, the end effector of the mechanical arm covers a spherical workspace with a radius of 6 m, and based on real-time path planning and force feedback control, safe and efficient garbage grabbing is realized. The garbage disposal production line unit includes a double-screw compressor (power 55 kW, adjustable pressure 0-200 MPa), a pyrolysis incinerator (secondary combustion chamber design), an SCR denitration system (catalyst V2O5-WO3 / TiO2), and an MBR membrane bioreactor (pore size 0.1 μm). The double-screw compressor directly receives the garbage unloaded by the mechanical arm, the incinerator is connected with the sewage treatment unit through a closed pipeline, and garbage volume reduction, harmless treatment and resource recycling are realized.

[0023] In the embodiments of the present application, the multi-source energy supply module is used to provide power for each module, including but not limited to wireless charging units (such as wireless charging piles), power supply through photovoltaic (monocrystalline silicon photovoltaic panels (200 square meters)), ocean wave energy (vertical axis wave power generator-diameter 3m, rated power 50kW per unit), battery energy storage (lithium titanate battery pack), power grid and diesel generator, and provides power for the wireless charging pile to charge the ship. Among them, the photovoltaic panel covers the platform roof, the wave power generator is arranged on the wave-encountering surface of the platform periphery, and the wireless charging coil of the wireless charging pile is embedded in the parking lot ground.

[0024] In the embodiments of the present application, the central control module has an edge computing server (1), an industrial switch (1), and an HMI touch screen (1), which are connected with each other through optical fibers, and a ROS (Robot Operating System) framework is used to realize multi-device collaborative control, and the communication delay is less than or equal to 10ms. Among them, the data of each module can be transmitted to the central control module for analysis and processing, and the results of the analysis and processing and the corresponding production instructions can be sent to each module for execution. In some embodiments, each module can also be configured with a separate processing unit to execute tasks, and each processing unit is connected with the central control module through optical fibers.

[0025] In one embodiment, the automatic unloading module is used to perform unloading tasks, specifically: collecting ship information of the ship transportation module, identifying the ship information to determine the position of the marine garbage and the position of the obstacle, and unloading the marine garbage to the garbage processing production line unit according to the position of the marine garbage and the position of the obstacle.

[0026] Among them, the ship information includes RGB images and obstacle data of the laser obstacle avoidance module, and a binocular vision sensor can be deployed or receive RGB images (resolution 1920x1080) collected by other modules of the binocular vision sensor, and then the image of the ship information is identified through a target detection network (such as YOLO) and binocular disparity, to determine the three-dimensional coordinate position of the marine garbage. For example, based on the YOLO target detection, the confidence , wherein, represents a weight coefficient, which is used to adjust the influence degree of IoU (the intersection over union of the prediction box and the real box) in the confidence calculation, represents a bias term (bias), which is used to shift and adjust the linear combination result before Sigmoid activation, the IoU threshold is greater than or equal to 0.5, the mAP@0.5 is 0.92, the IoU greater than or equal to 0.5 is used as the positive example judgment standard, the detection model has an accuracy of 0.92 under the mAP@0.5 evaluation, which guarantees the recognition accuracy and stability of the subsequent grabbing module of the unloading unit, accurately identifies the garbage area of the marine garbage, and then combines the binocular disparity wherein f is focal length, b is baseline distance (baseline distance in binocular vision is the horizontal distance between the optical centers of the left and right cameras), d is disparity, the binocular camera is fixedly installed on the ship body or the end of the mechanical arm, and the RGB images of the same garbage target are synchronously collected by the left and right cameras, and the pixel position difference of the target in the two images, i.e. the disparity, is extracted Disparity represents the horizontal pixel offset of the same target in the left and right images, reflecting the distance relationship between the target and the camera, combined with the focal length of the camera and the baseline distance between the left and right cameras , the formula is used to calculate the front and back distance of the target in the camera coordinate system , and the left and right offset X and the up and down offset Y are obtained by image pixel coordinate conversion, so as to determine the three-dimensional coordinate position (x, y, z) of the marine garbage, which describes the three-dimensional coordinate position of the marine garbage relative to the position of the camera optical center, which is equivalent to the center of gravity of the marine garbage, for example, the three-dimensional coordinate position of a certain plastic bottle is (0.25 m, -0.10 m, 1.80 m), i.e. located at the right side of the camera 25 cm, below the optical axis 10 cm, in front of 1.8 m. In addition, based on the obstacle data of the ship information, the position of the identified obstacle can be known.

[0027] Finally, after knowing the position of the marine garbage and the position of the obstacle, the marine garbage can be unloaded to the garbage disposal production line unit according to the three-dimensional coordinate position of the marine garbage and the position of the obstacle. Specifically: 1. According to the path planning algorithm, the three-dimensional coordinate position and the position of the obstacle, the moving path of the mechanical arm is determined.

[0028] Optionally, the path planning algorithm adopts an improved RRT* (rapidly-exploring random tree star) algorithm, an obstacle avoidance cost function is introduced, and the path length and obstacle avoidance distance are considered comprehensively to determine the moving path of the mechanical arm. Wherein, the obstacle avoidance cost function :

[0029] wherein n represents a certain moving path, , is the kth and k-1th discrete node in the path, is the obstacle avoidance weight, represents the minimum distance between the path and the nearest obstacle, which is from the obstacle position of the laser obstacle avoidance module or the laser radar point cloud, and this function can effectively guide the sampling tree to avoid high-risk areas and generate a safer and smoother mechanical arm movement path; B-spline smoothing: control point spacing 10 cm, curvature constraint The three-dimensional coordinate position needs to be considered in the discrete node determination process.

[0030] 2. Determine the control parameters of the mechanical arm according to the three-dimensional coordinate position and the mechanical parameters of the mechanical arm. The control parameters include angle, target clamping force, and control instruction.

[0031] First, according to the three-dimensional coordinate position (including x, y), the length of the connecting rod of the shoulder joint in the mechanical arm and the length of the connecting rod of the elbow joint , the angles of the shoulder joint and the elbow joint and the corresponding connecting rods are calculated respectively using a two-dimensional plane inverse kinematics model. The control signal of the angle can be sent to the servo system to drive the actuator to move accurately. The two-dimensional plane inverse kinematics model is as follows:

[0032]

[0033] The angle of the shoulder joint of the mechanical arm is , and the angle of the elbow joint is .

[0034] Then, when the end effector of the mechanical arm clamps the marine garbage according to the angle and the three-dimensional coordinate position, the contact area A of the end effector, the actual attitude and the actual speed are determined, and the target clamping force is determined according to the contact area, the preset friction coefficient and the preset garbage compression strength . Specifically, the clamping force model is introduced as follows:

[0035] wherein is the rubber-plastic friction coefficient, N is the normal clamping force, , A is the clamping contact area. This model effectively realizes force control grabbing and improves the stability and safety of the system. In the embodiments of the present application, a clamping force model based on double constraints of friction and material strength is introduced, which is suitable for various garbage material characteristics and prevents excessive damage during clamping.

[0036] Finally, according to the preset impedance stiffness coefficient , the preset impedance damping coefficient , the preset expected attitude , the preset expected speed , the actual attitude and the actual speed , the control instruction of the end effector is determined through the impedance control model. Specifically, the impedance control model is as follows:

[0037] wherein, the control instruction of the moment of force, ; by introducing impedance control strategy, according to the real-time feedback of six-dimensional force sensor, the compliance and tracking ability of the end effector are dynamically adjusted, the control instruction is generated based on the error between the expected / actual position and speed, and the end is ensured to realize smooth and accurate path tracking while ensuring that the garbage is not damaged.

[0038] 3. According to the control parameters and the moving path, the marine garbage is unloaded to the garbage processing production line unit by the mechanical arm.

[0039] In an embodiment, the garbage processing production line unit processes the marine garbage by multi-stage process to realize garbage volume reduction and harmlessness, through compression→incineration→wastewater treatment, the product of the previous link is input to the next link, specifically: 1. The marine garbage is compressed by a screw compressor, the volume is reduced to 1 / 5, the density is ≥1.2t / m³, and in the process of compression, the theoretical pressure distribution is determined by the garbage compression model , and the screw speed is controlled according to the theoretical pressure distribution and the PID control algorithm. Wherein, the garbage compression model calculates the theoretical pressure distribution at any position of the compression section by real-time sensing and modeling of parameters such as friction coefficient, screw diameter, speed and feeding rate , which provides pressure prediction basis for the compressor, dynamically adjusts the screw speed with PID control algorithm, realizes high pressure and low consumption, and stable compression ratio forming control, meets the forming needs of different garbage materials in the volume reduction process, and the screw speed is adjusted by PID, the error is <2%, realizes pressure closed loop control. Specifically, the garbage compression model:

[0040] wherein, is the theoretical pressure distribution, which specifically represents the pressure of the garbage when it is compressed to a certain position in the compression section, is the theoretical limit pressure in the screw terminal compression cavity, which is determined by the rated capacity of the compressor, is the current compression position (distance from the screw inlet), and k is the compression coefficient, reflecting the compression strength; is the friction coefficient (between the material and the screw cylinder wall), which is related to the type of garbage and the material of the cylinder wall, generally in 0.2~0.4, set according to the measured or table, which is 0.3 here; D refers to the diameter of the screw cross section, which is the geometric size of the equipment, , is the screw speed, such as 30rpm, which is set by the servo system or controlled by the sensor feedback. The feed volumetric flow rate (i.e., the volume of waste input per unit time) is obtained from a weighing and volume measurement system, or it can be estimated by the feed inlet size and rotation speed. .

[0041] 2. Compressed marine debris is incinerated in a pyrolysis incinerator to achieve NOx emissions ≤50mg / m³ and thermal efficiency >90%. Specifically, the combustion chamber is simulated using a Computational Fluid Dynamics (CFD) model (Realizable turbulence model), and the chemical reaction is simulated using an Eddy Dissipation Concept (EDC) model to simulate volatile matter combustion. The denitrification efficiency of SCR is... ,in, SCR entry NO x Concentration, measured by a flue gas analyzer. NO for SCR exports x Concentration, obtained from the online flue gas monitoring system (CEMS), The denitration reaction rate constant is related to the catalyst type and temperature (300~400°C), and can be obtained by fitting the reaction kinetics experimentally or by referring to tables in the literature. The average residence time of flue gas in the catalytic reactor is expressed by the following formula: Here, 2S is used, and V is the volume of the SCR reactor, which is determined by the structural dimensions of the equipment. This refers to the flue gas flow rate, either measured or designed.

[0042] 3. Wastewater generated during compression and incineration is filtered using an MBR membrane bioreactor, with a target COD ≤ 30 mg / L and membrane flux ≥ 15 LMH. MBR membrane filtration model:

[0043] in, Membrane bulk resistance / membrane resistance , The fouling resistance is updated in real time, and J is the membrane flux. Pressure difference across the membrane (transmembrane pressure difference) (The viscosity of the wastewater); among which, the backwashing strategy is configured as follows: backwash once every 2 hours, pressure 0.3MPa, duration 30 seconds.

[0044] In an embodiment, the multi-source energy supply module realizes energy self-sufficiency and reduces dependence on the power grid through hybrid energy scheduling and efficient wireless transmission. The multi-source energy supply module is specifically used to obtain a historical load power sequence of a preset time interval and task parameters of all tasks to be performed, predict demand power after the preset time interval according to the historical load power sequence and the task parameters, and adjust output power of various energy sources according to the demand power and priorities of the various energy sources.

[0045] wherein the preset time interval can be 1 hour / 60 minutes, the historical sequence of load power of the past 60 minutes is obtained and the task parameters of all tasks to be performed in the future 60 minutes , t is the current time, is based on the task parameters required to perform all tasks (such as at least one of the tasks of unloading, garbage disposal, and intelligent parking of the intelligent parking module), such as but not limited to parameters including path length, garbage density, initial extension amount, control parameters of hydraulic buffer piles, etc. A lookup table can be established in advance, and specific path length, garbage density, initial extension amount, and control parameters (range) of hydraulic buffer piles correspond to certain values, and then the sum is obtained , which is not specifically limited.

[0046] In an embodiment, the demand power after the preset time interval is predicted according to the historical load power sequence and the task parameters, specifically: the ambient temperature is obtained , the historical load power sequence , the task parameters and the ambient temperature are input into an LSTM network to predict the demand power after 60 minutes :

[0047] wherein the load prediction model based on LSTM is used to improve the feedforward regulation capability of hybrid energy scheduling, the network structure is 2-layer LSTM (128 units), and the output layer is linearly activated to improve the accuracy and prediction precision of energy scheduling, and to ensure that in each scheduling period, the future load demand is optimized and scheduled. The model input includes: the historical load power data of the past 60 minutes, provided by the BMS real-time monitoring system; the current ambient temperature, collected by the temperature sensor deployed on the platform.

[0048] In an embodiment, the output power of various energy sources is adjusted according to the demand power and the priorities of the various energy sources, including: 1. Construct a power generation cost function based on the power generation cost of various energy sources (such as unit power purchase cost, unit fuel cost, unit charge and discharge wear cost, etc.), the absolute value of the battery charge and discharge power, and the corresponding start state value of whether the diesel generator is started, with the specific goal of minimizing the power generation cost function:

[0049] wherein, is the unit power purchase cost of the mains, is the power obtained from the power grid, is the unit fuel cost of diesel power generation, is the start value of whether the diesel generator is started in the current hour (0 or 1), 1 representing start, is the unit charge and discharge wear cost of the battery energy storage, is the absolute value of the current battery charge and discharge power.

[0050] It should be noted that the minimization of the power generation cost function does not explicitly include the marine energy term, because the marginal cost of marine energy is approximately zero. Wave energy, tidal energy and other marine energy generation rely on natural ocean power and do not require fuel input. The operating cost mainly comes from fixed expenses such as equipment depreciation, maintenance and corrosion maintenance, which have no direct relationship with real-time power generation and do not increase with the increase of output. In addition, the output of marine energy is determined by the sea conditions and can be regarded as an exogenous variable. The system scheduling stage only determines its utilization or energy abandonment state, and will not generate additional costs due to the generation of one kilowatt-hour.

[0051] 2. Based on the constraint conditions and the priority of various energy sources, the output power of various energy sources is adjusted based on the column generation method to minimize the function value of the power generation cost function.

[0052] Optionally, the constraint conditions include: (1) Power balance, i.e. the sum of the final output power of various energy sources is greater than or equal to the demand power:

[0053] is the photovoltaic power, is the wave power (referred to as wave power), is the energy storage battery power, is the mains power, is the diesel power, is the demand power.

[0054] (2) Energy storage limit, i.e. the total power of the multi-source energy supply module is within a predetermined power range: . (3) Climbing constraint, i.e. the change rate of the wave power of marine wave energy is within a predetermined change range:

[0055] wherein, is the rated power of the wave power, is the wave power generation power at the current time t, and t-1 is the previous time.

[0056] It should be noted that the priorities of the energy sources from high to low are photovoltaic, ocean wave energy, battery energy storage, power grid, and diesel generator, and the constraint conditions include that the sum of the final output powers of various energy sources is greater than or equal to the demand power, the amount of power of the total power supply of the multi-source energy supply module is within a preset amount of power range, and the change rate of the wave power of the ocean wave energy is within a preset change range.

[0057] Therefore, based on the constraint conditions and the priorities of the various energy sources, based on the column generation method, the function value of the power generation cost function is minimized as the goal, and finally the output powers of the various energy sources (including whether to start the diesel generator) are obtained, the energy supply mode of different energy sources and the specific energy supply output power are dynamically adjusted, and the coordinated operation of multiple energy sources is effectively realized.

[0058] In an embodiment, the wireless charging unit transmits 50kW power with an efficiency of ≥85%, with a tolerance of ±15cm, and is specifically used for: 1. When adjusting the output powers of various energy sources to charge the ship transportation module, wireless power transmission is performed through magnetic resonance coupling technology.

[0059] wherein, the SOC of the ship is monitored in real time and the load demand is predicted, the SOC detection is performed by reading the BMS data on the ship through the CAN bus, the wireless charging condition is triggered, such as whether the charging flag is 1 or 0, 1 triggers the wireless charging unit to charge the ship, and the wireless power transmission is performed using the magnetic resonance coupling technology during charging.

[0060]

[0061] 2. During the process of power transmission, the resonance frequency is kept within a preset frequency range by dynamically adjusting the resonance capacitor.

[0062] The embodiment of the present application introduces a magnetic resonance coupling model:

[0063] wherein, the coupling efficiency is determined by the magnetic coupling coefficient (0.6, double D coil design), the quality factor of the transmitting coil and the receiving coil . The system calculates the transmitting power through the model according to the allocated target energy supply power, and dynamically adjusts the resonance capacitor to maintain the resonance frequency constant within a preset frequency range (for example, around 85khz), to realize offset compensation, winding of the wire, ).

[0064] 3. Determine the transmission efficiency of wireless power transmission, if the transmission efficiency is less than the efficiency threshold (for example, 85%), exclude the current output power of various energy sources, return to the step of adjusting the output power of various energy sources based on the constraints and the priority of various energy sources, that is, readjust the combined power supply scheme, based on the column generation method to minimize the function value of the power generation cost function, until the transmission efficiency is greater than or equal to the efficiency threshold, to ensure the charging efficiency and energy self-sufficiency.

[0065] In an embodiment, the intelligent docking module is responsible for guiding, positioning and stabilizing the ship, in particular for: 1. Coarse positioning by ultra-wideband (UWB) to determine the first coordinate position of at least one ship in the ship transport module.

[0066] Optionally, communicate with the 4 parent port base stations through the ship's on-board tag, and use TDOA to calculate the initial first coordinate position of at least one ship.

[0067] 2. Obtain the velocity vector of the ship (including the velocity component in the berthing direction) through the inertial measurement unit, pre-adjust the hydraulic buffer pile through the first coordinate position and the velocity vector, and determine the pre-charging pressure value of the hydraulic buffer pile , that is, the pre-charging pressure value of the hydraulic buffer pile at time (moment) t:

[0068] Optionally, use time window prediction method to fit the berthing path in the next 10-30 seconds, determine whether to trigger the pre-adjustment process of the hydraulic buffer pile based on whether the first coordinate position enters the pre-defined buffer zone and whether the approach angle is less than 20° with the platform normal, to determine the pre-charging pressure value. Wherein, is the equivalent mass of the ship (considering the load and added mass), is the velocity component of the ship in the berthing direction, is the expected buffer stroke, is the number of buffer piles, is the effective area of the hydraulic cylinder piston of a single buffer pile, is the safety factor, is the maximum pre-charging pressure allowed by the system (preset equipment constraint value).

[0069] 3. Obtain the point cloud data of the ship through the laser radar unit, pre-load the hydraulic buffer pile through the point cloud data, determine the initial extension amount of the hydraulic buffer pile, and control the hydraulic buffer pile to enter the buffer preparation state according to the pre-charging pressure value and the initial extension amount.

[0070] Since the above is a coarse positioning, it is necessary to further improve the positioning accuracy to ±2cm and correct the ship's attitude. Therefore, point cloud data is used to preload the hydraulic buffer piles and determine the initial expansion and contraction of the hydraulic buffer piles, including: First, through point cloud data and preset template point clouds (i.e., point cloud data of a ship approaching its standard berthing state), the rotation and translation parameters of the hydraulic buffer piles are iteratively processed to determine the minimum registration error:

[0071] Where i is the point number. For the number of points, , The rotation and translation parameters to be optimized are: This is a curvature-based weighting function (used to mitigate the influence of edges / outliers). This is a regularization factor used to suppress large-scale translation errors. Point cloud curvature based on the i-th point The weight, : Regularization coefficient, suppressing abnormal translation. Additionally, the iterative steps are: voxel filtering downsampling (5cm resolution); FPFH feature matching (feature dimension 33); Levenberg-Marquardt optimization (50 iterations, convergence threshold 1e-6).

[0072] Secondly, the initial extension / retraction amount is determined by multiplying the minimum registration error by the arm length of the hydraulic buffer pile.

[0073] Optionally, based on the minimum registration error finally obtained in the previous iteration... (Or posture deviation) and the arm length of the hydraulic buffer pile The initial expansion / contraction is determined by multiplying the product of (e.g., 2m) by the product of (e.g., 2m). , It should be noted that this initial expansion is used to preload the buffer piles, so as to apply a directional flexible corrective force to the ship during the actual berthing process, thereby completing the gradual correction of attitude during physical contact.

[0074] 4. During the contact process between the ship and the hydraulic buffer pile, determine the positional deviation between the ship and the hydraulic buffer pile in real time. (i.e., the position deviation at time t, obtained through LiDAR ranging or ship attitude transformation, which can utilize existing methods) and the rate of change of the position deviation. ( (used to indicate whether the error is increasing or decreasing), based on positional deviation. rate of change ( and PID algorithm, real-time output the control parameters of the hydraulic buffer pile.

[0075] It should be noted that during the process of the ship contacting the hydraulic buffer pile, the target is to absorb the impact energy of the ship, and the position error is less than 5 cm. Among them, according to the position deviation , the rate of change ( ) and PID algorithm, real-time output the control parameters of the hydraulic buffer pile, including: Firstly, the position deviation and the rate of change are mapped to a preset number of fuzzy sets as input variables.

[0076] In the embodiment of the application, a dynamic model of the hydraulic buffer pile is provided:

[0077] Among them, is the equivalent mass of the hydraulic buffer pile, such as 500 kg (including the hydraulic cylinder assembly, the equivalent mass of a single hydraulic buffer pile system along the impact direction), the variable represents the dynamic deformation displacement of the hydraulic buffer pile in the axial direction, represents the effective component of the ship impact force in the direction of the buffer pile, and the angle is obtained from the angle between the ship heading and the platform normal, and the ship contact buffer pile time is taken as the initial time , the differential equation is solved in the contact stage (such as 120-300 seconds), and then the evaluation and control of the buffer energy absorption efficiency are realized, c is the damping coefficient (calculated by the throttle hole diameter , is the equivalent stiffness, and .

[0078] In the embodiment of the application, in the dynamic buffering stage, the fuzzy PID control algorithm is used to adjust the response characteristics of the hydraulic buffer on line, the PID gain coefficient is dynamically adjusted through the fuzzy inference rule, and the adaptive control of the buffer response force is realized. Specifically, the position error and the error change rate are mapped to a preset number (for example, 7) of fuzzy sets (NB to PB) as input variables.

[0079] Secondly, the input variables are variable fuzzed through a preset number of series membership functions, and according to the preset rules corresponding to the variable fuzzed input variables and the control parameters, the control parameters of the hydraulic buffer pile are calculated in real time.

[0080] Specifically, the input variables are fuzzed by using 7-level membership functions, and the rule base includes 49 preset rules of the "IF-THEN" type corresponding to the variable fuzzed input variables and the control parameters:

[0081] Then, based on the formula solution output hydraulic buffer pile control parameters , used to control the orifice, overflow valve or solenoid valve operating parameters, and then realize the impact absorption and attitude stability control:

[0082] Wherein, is the position error, that is, the deviation of the actual position of the ship and the center position of the buffer, error rate (i.e. displacement error first derivative, reflecting the approach trend), PID controller initial proportion, integral, differential coefficient, Incremental term obtained by fuzzy logic dynamic adjustment (online compensation for initial value), initial parameters ; Wherein, NB is Negative Big, NM is Negative Medium, NS is Negative Small, ZO is Zero, PS is Positive Small, PM is Positive Medium, and PB is Positive Big.

[0083] In an embodiment, the central control module realizes a sampling rate of 1 kHz, a delay of ≤10 ms, TSN network scheduling: cycle 1 ms, time slot allocation: sensor data (40%), control command (30%), state query (20%), redundancy (10%); Data compression: ZigZag encoding (compression rate ≥50%). The central control module is used to: 1. Obtain the state information sequence of the ship transportation module, the garbage disposal module, the multi-source energy supply module and the intelligent docking module.

[0084] Optionally, the working state of all components / devices in each module at each time can be collected by the central control module to form a state information sequence. For example, the working state of the ship in the ship transportation module, the working state of the mechanical arm, the double screw compressor, the SCR denitration system, the pyrolysis incinerator MBR membrane bioreactor, etc. in the garbage disposal module, the working state of each device providing energy in the multi-source energy supply module, and the working state of the hydraulic buffer pile in the intelligent docking module.

[0085] 2. According to the state information sequence and the LSTM model with fusion attention mechanism (i.e. LSTM-Attention model), output the fault index. Wherein, the LSTM-Attention model:

[0086] Wherein, the influence weight of the final prediction for the moment t, is the output (hidden state) of the LSTM model at time step t, is the global attention vector learned by the model. The hidden state (i.e., feature representation) output by the LSTM at time step t is obtained by step-by-step encoding of the input state information sequence, which represents the comprehensive information at that time point. The global attention vector (trainable parameter) learned by the model is used to represent the "feature direction" that the model pays attention to, and is adjusted during the training process. The attention weight at time step t represents the importance of the moment in the final output.

[0087] Then, the fault index is calculated by the formula :

[0088] wherein, is the weight matrix of the fully connected layer, and b is the bias term, is the activation function. The fault index is predicted from the state information sequence by the LSTM+Attention model. The final fault index is based on the LSTM hidden state and the weighted output of the Attention mechanism, and is mapped by a fully connected layer. The fault index .

[0089] 3. When the fault index is greater than or equal to the fault threshold, a fault is prompted and / or the redundant device is switched.

[0090] Optionally, the fault threshold is 0.8, and if is greater than or equal to the fault threshold, it indicates that there is a fault, the specific fault device is determined, and the fault is prompted and / or the redundant device is switched.

[0091] In an embodiment, the central control module can also be used for human-computer interaction and remote maintenance, and digital twinning (Unity3D constructs a three-dimensional visualization model with a synchronization delay of <200ms) and remote monitoring and parameter adjustment can be realized through the provided Web interface (based on the Django framework, which displays the device status in real time).

[0092] The system of the embodiment of the present application has at least the following advantages: 1. The system solves the problems of low docking efficiency, inconvenient energy supply, manual unloading, and serious environmental pollution in traditional garbage disposal systems, realizes intelligent operation of the whole process of garbage collection, transportation, and disposal, significantly improves the automation level and environmental friendliness of garbage disposal, provides efficient infrastructure support for marine ecological protection, urban solid waste management, and carbon neutralization goals, and has significant economic comprehensive benefits.

[0093] 2、Combining multi-sensor fusion positioning technology, significantly shortening the ship docking time, improving positioning accuracy, and reducing the operation complexity of traditional anchoring method.

[0094] 3、Through renewable energy priority scheduling and intelligent wireless charging technology, the charging efficiency is greatly improved, and the dependence on external power grid is reduced.

[0095] 4、Based on advanced path planning and mechanical arm cooperative control technology, realize the whole process automation of garbage transfer, significantly reduce the labor cost and operation risk. At the same time, through the multi-level linkage of compression and volume reduction, pyrolysis incineration and sewage treatment, effectively reduce the volume of garbage and pollutant emissions, and improve environmental friendliness.

[0096] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0097] In addition, the terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0098] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any skilled person in the art can easily think of various changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A system of offshore platform for treating marine litter, characterized in that, The method comprises the following steps: a ship transportation module is used to collect marine garbage; a garbage disposal module comprises an automatic unloading unit and a garbage disposal production line unit, the automatic unloading module is used to perform an unloading task, collect ship information of the ship transportation module, identify the ship information to determine the position of marine garbage and the position of obstacles, unload marine garbage to the garbage disposal production line unit according to the position of marine garbage and the position of obstacles, and the garbage disposal production line unit is used to perform a garbage disposal task and dispose marine garbage; a multi-source energy supply module is used to obtain a historical load power sequence of a preset time interval and task parameters of all tasks to be performed, predict a demand power after the preset time interval according to the historical load power sequence and the task parameters, and adjust the output power of various energy sources according to the demand power and the priority of various energy sources.

2. The offshore platform system for processing marine litter according to claim 1, characterized in that: The step of identifying the ship information to determine the position of marine garbage and the position of obstacles and unloading marine garbage to the garbage disposal production line unit according to the position of marine garbage and the position of obstacles comprises the following steps: image recognition of the ship information is performed by a target detection network and a binocular disparity, three-dimensional coordinate positions of marine garbage are determined, and obstacle positions are identified according to obstacle data of the ship information; a moving path of a mechanical arm is determined according to a path planning algorithm, the three-dimensional coordinate positions and the obstacle positions; control parameters of the mechanical arm are determined according to the three-dimensional coordinate positions and mechanical parameters of the mechanical arm; marine garbage is unloaded to the garbage disposal production line unit by the mechanical arm according to the control parameters and the moving path.

3. The offshore platform system for processing marine litter according to claim 2, characterized in that: The step of determining the control parameters of the mechanical arm according to the three-dimensional coordinate positions and the mechanical parameters of the mechanical arm comprises the following steps: angles of shoulder joints and elbow joints and corresponding connecting rods are calculated respectively by using a two-dimensional plane inverse kinematics model according to the three-dimensional coordinate positions, a connecting rod length of the shoulder joints and a connecting rod length of the elbow joints; a contact area, an actual posture and an actual speed of an end effector of the mechanical arm are determined when the end effector clamps marine garbage according to the angles and the three-dimensional coordinate positions, and a target clamping force is determined according to the contact area, a preset friction coefficient and a preset garbage compression strength; a control instruction of the end effector is determined by an impedance control model according to a preset impedance stiffness coefficient, a preset impedance damping coefficient, a preset expected posture, a preset expected speed, the actual posture and the actual speed of the end effector; the control parameters comprise the angles, the target clamping force and the control instruction.

4. The offshore platform system for processing marine litter according to claim 1, characterized in that: The step of disposing marine garbage comprises the following steps: marine garbage is compressed by a screw compressor, a theoretical pressure distribution is determined by a garbage compression model in the process of compression, and screw rotation speed is controlled according to the theoretical pressure distribution and a PID control algorithm; compressed marine garbage is incinerated by a pyrolysis incinerator; wastewater generated in the compression and incineration processes is filtered by an MBR membrane bioreactor.

5. The offshore platform system for processing marine litter according to claim 1, characterized in that: The energy sources include photovoltaic, ocean wave energy, battery energy storage, power grid and diesel generator; and the adjusting the output power of various energy sources according to the demand power and the priority of various energy sources comprises: constructing a power generation cost function according to the power generation cost of various energy sources, the absolute value of battery charge and discharge power and the starting state value corresponding to whether the diesel generator is started; adjusting the output power of various energy sources based on column generation method to minimize the function value of the power generation cost function based on the constraint conditions and the priority of various energy sources; wherein the priority of energy sources from high to low is photovoltaic, ocean wave energy, battery energy storage, power grid and diesel generator, and the constraint conditions include that the sum of the final output power of various energy sources is greater than or equal to the demand power, the electric quantity of the total power supply of the multi-source energy supply module is in a preset electric quantity range, and the change rate of the wave power of the ocean wave energy is in a preset change range.

6. The offshore platform system for processing marine litter according to claim 5, characterized in that: The multi-source energy supply module further comprises a wireless charging unit, and the wireless charging unit is used for: wirelessly transmitting electric energy by magnetic resonance coupling technology when the output power of various energy sources is adjusted for charging the ship transportation module; in the process of electric energy transmission, the resonance frequency is kept within a preset frequency range by dynamically adjusting the resonance capacitance; determining the transmission efficiency of wireless electric energy transmission, if the transmission efficiency is less than the efficiency threshold, excluding the current output power of various energy sources, returning to the step of adjusting the output power of various energy sources based on column generation method to minimize the function value of the power generation cost function based on the constraint conditions and the priority of various energy sources, until the transmission efficiency is greater than or equal to the efficiency threshold.

7. The offshore platform system for processing marine litter according to any of claims 1-6, characterized in that: The offshore platform system for processing marine garbage further comprises an intelligent docking module, and the intelligent docking module is used for: coarse positioning by ultra-wideband to determine the first coordinate position of at least one ship in the ship transportation module; obtaining the velocity vector of the ship by an inertial measurement unit, pre-adjusting the hydraulic cushion pile by the first coordinate position and the velocity vector to determine the pre-charging pressure value of the hydraulic cushion pile; obtaining the point cloud data of the ship by a laser radar unit, pre-loading the hydraulic cushion pile by the point cloud data to determine the initial extension amount of the hydraulic cushion pile, and controlling the hydraulic cushion pile to enter the buffering preparation state according to the pre-charging pressure value and the initial extension amount; determining the position deviation and the change rate of the position deviation between the ship and the hydraulic cushion pile in real time during the contact between the ship and the hydraulic cushion pile, and outputting the control parameters of the hydraulic cushion pile in real time according to the position deviation, the change rate and the PID algorithm.

8. The offshore platform system for processing marine litter according to claim 7, characterized in that: The pre-loading of the hydraulic cushion pile by the point cloud data to determine the initial extension amount of the hydraulic cushion pile comprises: iteratively processing the rotation parameters and the translation parameters of the hydraulic cushion pile by the point cloud data and a preset template point cloud to determine the minimum registration error; determining the initial extension amount according to the product of the minimum registration error and the arm length of the hydraulic cushion pile.

9. The offshore platform system for processing marine litter according to claim 7, characterized in that: The outputting of the control parameters of the hydraulic cushion pile in real time according to the position deviation, the change rate and the PID algorithm comprises: Map the position deviation and the change rate to a preset number of fuzzy sets as input variables; Variable fuzzification is performed on the input variables by a preset number of series membership functions, and the control parameters of the hydraulic cushion pile are calculated in real time according to a preset rule corresponding to the variable-fuzzified input variables and the control parameters.

10. The offshore platform system for processing marine litter according to claim 7, characterized in that: The offshore platform system for processing marine garbage further comprises a central control module, which is configured to: acquire state information sequences of the ship transportation module, the garbage processing module, the multi-source energy supply module and the intelligent docking module; output a fault index according to the state information sequences and an LSTM model with a fusion attention mechanism; prompt a fault and / or switch a redundant device when the fault index is greater than or equal to a fault threshold.

Images