Intelligent Decision-Making Optimization Method for Dredging Operations Based on Multi-Source Data Fusion

Abstract

This invention relates to the field of dredging construction technology, specifically disclosing an intelligent decision-making optimization method for dredging operations based on multi-source data fusion. The method includes collecting and preprocessing multi-source data, which includes general data and vessel-specific data for different dredging vessel types. Based on the dredging vessel type, a vessel-specific dual-drive coupling sub-model is constructed and executed to fuse the multi-source data, and vessel-specific spectral coupling analysis and active anomaly identification are performed. This invention achieves a combination of unified framework and vessel-specific approach by constructing a vessel-specific dual-drive coupling sub-model and data system, enabling the decision-making system to accurately match the unique physical mechanisms of specific vessel types while operating within a unified intelligent framework.

Description

Technical Field

[0001] This invention relates to the field of dredging construction technology, and in particular to an intelligent decision-making optimization method for dredging operations based on multi-source data fusion. Background Technology

[0002] Dredging operations are crucial for ensuring navigation and land reclamation, and their operational decisions are paramount in the complex marine environment. Traditional dredging relies heavily on human experience or static construction plans, which struggle to adapt to dynamic environmental changes in real time, resulting in low efficiency and excessive energy consumption. To address this, the industry has been seeking to introduce intelligent decision-making models, but in practice, this immediately encounters an irreconcilable conflict between versatility and precision: different types of dredging vessels (such as trailing suction hoppers and cutter suction hoppers) have drastically different physical characteristics and operating mechanisms, requiring an accurate decision-making model deeply tied to the specific characteristics of each vessel type; however, an overly specialized model loses its versatility and portability. Existing technologies either employ a one-size-fits-all universal model, leading to inaccurate decisions due to a lack of vessel type specificity, or use highly customized single-vessel models, resulting in high development costs and failing to establish a unified intelligent decision-making framework.

[0003] To make the model intelligent, it must be able to automatically evolve based on real-time multi-source data, achieving bidirectional driving and fusion of data and model. However, this ingeniously designed adaptive capability also has its limitations:

[0004] On the one hand, when the system attempts to respond to changes in the real-time environment by dynamically adjusting weights, such as prioritizing energy consumption or schedule, if the environmental data fluctuates near the decision threshold, this high responsiveness will immediately cause the system's control commands to oscillate at high frequency between the two strategies, resulting in equipment wear or even control failure, making responsiveness and stability mutually exclusive.

[0005] On the other hand, when the system attempts to automatically adjust its parameters based on the deviation between the model output and the sensor data, if it encounters sensor failure or extreme conditions, the adaptive mechanism may mistakenly treat these dirty data or abnormal deviations as features that the model needs to learn, and then incorrectly adjust the parameters to cater to these erroneous data, ultimately leading to the contamination or even collapse of the entire decision model. Summary of the Invention

[0006] This invention aims to at least partially solve one of the technical problems in related technologies. Therefore, the objective of this invention is to propose an intelligent decision-making optimization method for dredging operations based on multi-source data fusion, in order to achieve stable and efficient intelligent dredging operations.

[0007] To achieve the above objectives, a first aspect of the present invention proposes an intelligent decision-making optimization method for dredging operations based on multi-source data fusion, comprising the following steps:

[0008] Collect and preprocess multi-source data, including general data and vessel-specific data for different dredging vessel types;

[0009] Based on the dredging vessel type, a vessel-specific dual-drive coupling sub-model is constructed and executed to fuse the multi-source data and perform vessel-specific spectral coupling analysis and vessel-specific active anomaly identification.

[0010] Based on the results of the multi-source data fusion, ship-specific dynamic weight adjustment and ship-specific safety constraints are performed, and ship-specific pre-trained working condition library is called to generate dredging operation decisions.

[0011] The effectiveness of the dredging operation decisions is monitored and verified using ship-specific resilience, and a ship-specific tamper-proof report is generated.

[0012] Based on the execution results, the ship-specific dual-drive coupling sub-model is iteratively optimized.

[0013] To achieve the above objectives, a second aspect of the present invention proposes an intelligent decision-making optimization system for dredging operations based on multi-source data fusion, comprising:

[0014] The data acquisition module is used to collect and preprocess multi-source data, which includes general data and ship-specific data for different dredging vessel types.

[0015] The model execution module is used to construct and execute a ship-specific dual-drive coupling sub-model based on the dredging vessel type, fuse the multi-source data, and perform ship-specific spectral coupling analysis and ship-specific active anomaly identification.

[0016] The decision generation module is used to perform ship-type-specific dynamic weight adjustment and ship-type-specific safety constraints based on the results of the multi-source data fusion, and call the ship-type-specific pre-trained working condition library to generate dredging operation decisions.

[0017] The monitoring and verification module is used to monitor and verify the execution effect of the dredging operation decision on a ship type-specific resilience basis, and generate a ship type-specific anti-tampering report.

[0018] The iterative optimization module is used to iteratively optimize the ship-type-specific dual-drive coupling sub-model based on the execution results.

[0019] To achieve the above objectives, a third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory. When the computer program is executed by the processor, it implements the above-described intelligent decision optimization method for dredging operations based on multi-source data fusion.

[0020] The intelligent decision-making optimization method for dredging operations based on multi-source data fusion in this invention achieves a combination of unified framework and ship-specific design by constructing a dual-drive coupled sub-model and data system specific to each ship type. This allows the decision-making system to accurately match the unique physical mechanisms of a specific ship type while operating within a unified intelligent framework. More importantly, by introducing a strategy with hysteresis and smooth transition, this invention achieves a sensitive response to environmental changes while eliminating control command oscillations caused by data noise jitter to a certain extent. This balances the responsiveness and stability of the system, ensuring the smooth operation of physical equipment. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the intelligent decision-making optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0022] Figure 2 This is a schematic diagram of extracting wave peaks and troughs from a wave image sequence in the intelligent decision-making optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0023] Figure 3 This is a schematic diagram comparing the spectrum of equipment deviation and environmental disturbance in the intelligent decision-making optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0024] Figure 4 This is a schematic diagram of the average coupling strength factor variation curve in the intelligent decision optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0025] Figure 5 This is a schematic diagram of the construction process and index curve of the comprehensive wave hazard index in the intelligent decision optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0026] Figure 6 This is a schematic diagram comparing predicted and measured values ​​in the intelligent decision-making optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0027] Figure 7 This is a simulation diagram of the smooth transition function in the intelligent decision optimization method for dredging operations based on multi-source data fusion provided by the present invention.

[0028] Figure 8 This is a schematic diagram of the operational structure of the intelligent decision-making optimization system for dredging operations based on multi-source data fusion provided by the present invention;

[0029] Figure 9 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0030] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0031] The following description, with reference to the accompanying drawings, outlines an intelligent decision-making optimization method, system, and electronic device for dredging operations based on multi-source data fusion, according to embodiments of the present invention.

[0032] Example 1:

[0033] This embodiment provides a complete implementation process for an intelligent decision-making optimization method for dredging operations based on multi-source data fusion. The method in this embodiment is typically executed by an intelligent decision-making optimization system for dredging operations deployed on a local or remote cloud server on the dredging vessel. The system's hardware may include: onboard multi-source data acquisition sensors, data acquisition cards, edge computing gateways, and satellite or 5G communication modules for data communication; its software includes a data acquisition module, a model execution module, a decision generation module, a monitoring and verification module, and an iterative optimization module.

[0034] like Figure 1 As shown, the complete method flow of this embodiment specifically includes the following five core steps:

[0035] Step S1: Acquisition and Preprocessing of Multi-Source Data

[0036] The first step is to collect all data related to dredging operations comprehensively and in real time. In order to achieve subsequent intelligent decision-making specific to the ship type, this invention divides the data sources into two categories: general data and ship type-specific data.

[0037] 1. Collection of General Data: General data refers to the basic environmental and status data that all types of dredging vessels need to refer to during operation. In this embodiment, the general data includes at least:

[0038] High-precision positioning data: Using differential global satellite navigation systems, such as RTK-GNSS, including GPS and BeiDou, to obtain the real-time three-dimensional coordinates (longitude, latitude, and elevation) and heading and speed information of the dredging vessel, this is the basis for operation path planning, earthwork volume calculation, and compliance inspection of the construction area.

[0039] Hydrogeomorphological data: Using shipborne single-beam or multi-beam echo sounders, underwater topographic and geomorphological data and water depth data of the operational area are acquired in real time. Simultaneously, real-time tide level data is obtained through connected tide gauges or the Tide-Link system, used to correct all measured water depths to a unified reference elevation.

[0040] Meteorological and hydrological data: Using shipboard automatic weather stations and current meters, information such as wind speed, wind direction, current speed, and current direction at the work site is obtained. This information is crucial for predicting ship drift and adjusting propeller power.

[0041] Ocean wave data: This is one of the key inputs of this invention. Traditional ocean wave data, such as that from buoys and shore-based radar, suffers from fixed locations and slow updates. In this embodiment, however, ocean wave data is acquired using a more advanced remote sensing measurement method based on unmanned aerial vehicles (UAVs). Specifically:

[0042] Operators can control a drone equipped with a high-definition or hyperspectral camera to hover or patrol over the dredging vessel's operating area, especially over the wave-facing side. The drone collects continuous wave image sequences and transmits the image data back to the decision-making system in real time.

[0043] The system's wave analysis module then processes the received image sequences. For example, it uses computer vision algorithms, such as optical flow, particle image velocimetry (PIV), or deep learning-based semantic segmentation networks, to identify wave peaks and troughs in the images and track their movement. In this way, the system can extract key, quantified wave parameters from the visual information, specifically including:

[0044] wave height : refers to the vertical distance between consecutive wave crests and troughs;

[0045] cycle : Refers to the time interval between two consecutive wave peaks (or troughs) passing through the same location;

[0046] Wave breaking height : Refers to the instantaneous height of the wave crest when the waves break in shallow water or due to wind force;

[0047] Surge water level This refers to the instantaneous, non-periodic rise in water level caused by storms or long-period waves, such as nearshore swells.

[0048] Surge duration This refers to the duration of the aforementioned surge in water level.

[0049] These high spatiotemporal resolution wave parameters, acquired in real time by drones, serve as the direct basis for subsequent calculations of the comprehensive wave hazard index, and their accuracy surpasses that of traditional methods.

[0050] like Figure 2 The diagram visually illustrates the extraction of wave crests and troughs from a sequence of ocean wave images. This involves processing ocean wave images continuously captured by drones over the dredging area to identify the location of each wave crest and trough, which then serves as the basis for calculating key parameters such as wave height, period, and hazard index.

[0051] Figure 2 The blue main curve simulates the changes in sea surface waveforms after image processing and time series reconstruction, representing the dynamic process of waves rising and falling within one cycle. Red dots mark the wave crests automatically identified by the system; these crests are the highest points in each complete wave, representing the maximum rise height of the wave. Green dots mark the corresponding wave troughs, the lowest points of the wave, reflecting the lowest point of wave energy.

[0052] pass Figure 2 The system can reliably identify the occurrence time and height of each wave crest and trough, facilitating further accurate calculation of wave height and period. This extracted high-precision wave height and period data will be directly used to calculate the wave hazard index, which in turn serves as the input variable for safety constraints in subsequent decision-making modules.

[0053] 2. Collection of ship-specific data: The system must collect key data that reflects the unique operating mechanism of a specific dredging vessel, such as a trailing suction hopper, cutter suction hopper, or grab bucket dredging vessel. This data is the cornerstone for building a ship-specific model.

[0054] (1) For trailing suction hopper dredging vessels (TSHD):

[0055] Mud hopper loading capacity: This is calculated in real time by an array of pressure sensors installed at the bottom of the mud hopper or by monitoring changes in the ship's draft. This data is directly related to the economic loading of the trailing suction hopper operation.

[0056] Sloshing amplitude of mud and sand in mud tank: By monitoring the level sensor or inclinometer installed on the inner wall of the mud tank, the violent sloshing of the mud and sand mixture in the tank can seriously affect the navigation stability of the ship, especially in bad sea conditions.

[0057] This also includes the vacuum level of the rake head, the speed and power of the mud pump, the drag force of the rake arm, and the turbidity of the overflow pipe.

[0058] (2) For CSD (Cut Suction Dredger):

[0059] Sludge discharge pipe pressure: Pressure sensors are installed at the sludge pump outlet and key nodes in the sludge discharge pipeline, such as above-water and underwater joints. This pressure is the driving force for maintaining the long-distance transport of sludge mixtures and is also a core indicator for determining whether the pipeline is blocked.

[0060] Sludge discharge flow rate fluctuation: The instantaneous flow rate is monitored using an electromagnetic flow meter or ultrasonic flow meter installed on the sludge discharge pipe. This embodiment focuses more on its fluctuation value, that is, the variance or standard deviation of the flow rate. Severe flow rate fluctuations usually indicate that the cutter head's digging conditions are unstable, such as encountering hard soil or rocks.

[0061] It also includes the cutter head's rotational speed and torque, digging depth, the position of the traverse trolley, traverse speed, anchor cable tension, etc.

[0062] (3) For grab-type dredging vessels:

[0063] Grab bucket opening / closing degree: Monitored by an angle sensor installed on the grab bucket hydraulic or cable system;

[0064] Grab bucket lifting speed: obtained by monitoring the encoder or speed sensor of the winch;

[0065] These two parameters are key to defining the efficiency of a grab bucket operation cycle. Other parameters include the weight of the grab bucket, and the ship's roll and pitch angles.

[0066] 3. Data Preprocessing: The raw data collected, whether general or specialized, is often disorganized and cannot be directly used in the model. Preprocessing steps aim to improve data quality and include:

[0067] Data cleaning: Identify and remove obvious outliers and null values ​​caused by sensor malfunctions or communication interruptions;

[0068] Noise filtering: Apply filters (such as Kalman filtering and moving average filtering) to high-frequency fluctuating signals, such as sludge discharge pipe pressure and sludge sloshing amplitude, to extract the true trend signal;

[0069] Timestamp alignment: Because different sensors, such as GPS, flow meters, and wave height meters, have different sampling frequencies and delays, all data must be unified to the same time base;

[0070] Data normalization: scaling data of different dimensions, such as pressure, flow rate, and rotational speed, to a uniform range (e.g., 0 to 1 or -1 to 1) to facilitate input into subsequent neural network and other models.

[0071] Step S2: Construct and execute the ship-specific dual-drive coupling sub-model

[0072] This step involves the system calling a dual-drive coupling sub-model that matches the current dredging vessel type, such as a cutter suction dredger. The dual-drive coupling sub-model is built on the basis of a dredging-specific physical model and performs in-depth fusion and analysis on the pre-processed multi-source data.

[0073] 1. The meaning of dual-drive coupling:

[0074] Mechanism-driven: The dredging-specific physical model incorporates physical and hydrodynamic formulas that reflect the operating principles of the vessel type, describing the sediment transport, hydrodynamic, and energy consumption characteristics of a specific vessel type. For example, for cutter suction dredgers, the model incorporates a formula for calculating the pressure loss of the two-phase flow of sediment mixture in the pipeline; for trailing suction dredgers, it incorporates a physical model of sediment settling in the silt chamber.

[0075] Data-driven: The model incorporates artificial intelligence models trained using historical and real-time data, such as deep neural networks and support vector machines. Specifically, it extracts and fuses feature parameters that are strongly correlated with dredging performance indicators from both vessel-specific and general data. These strongly correlated feature parameters include wave height and mud hopper loading. For example, a neural network is used to fit the nonlinear relationship between cutter torque and sludge discharge pipe pressure fluctuations under the same operating conditions.

[0076] Coupling: This refers to the fact that the mechanistic model and the data model do not operate independently, but rather mutually correct and couple with each other. The mechanistic model provides the physical constraints and boundaries for the data model, preventing the data model from making predictions that violate common sense; while the data model, based on real-time data, in turn corrects the empirical parameters in the mechanistic model that are difficult to measure.

[0077] 2. Ship type-specific proactive anomaly detection:

[0078] A key task during model execution is proactive anomaly identification. Traditional anomaly identification methods, such as simple threshold alarms, are very coarse. This invention employs an advanced diagnostic method based on spectral coupling analysis. The core idea of ​​this method is that, as a complex power system, the decline in the operational efficiency or the increase in safety risks of a dredging vessel is often not caused by a single factor, but rather by the resonance or strong coupling of environmental disturbances and equipment deviations at a specific frequency.

[0079] To quantify this degree of coupling, the system performs the following calculations:

[0080] A. The system acquires the principal component sequence of equipment deviation and the environmental disturbance sequence;

[0081] B. The system performs a Fourier transform (FFT) on these two time series to convert the time-domain signals into frequency-domain signals, obtaining their respective deviation power spectra. and perturbation power spectrum These two spectra show the energy at different frequencies. Distribution on;

[0082] C. Simultaneously, the system calculates the cross-power spectrum of these two signals. The cross-power spectrum is a complex number; the combination of its real and imaginary parts reflects the cross-power spectrum of two signals at a specific frequency. The intensity of correlation or coordinated motion;

[0083] D. The system defines one or more sensitive frequency bands based on the ship type, such as cutter suction hull, for example... arrive From 0.05Hz to 0.2Hz;

[0084] E. The system performs integral calculations within this sensitive frequency band: calculating the low-frequency cooperative response. The integral here is performed on a specific component of the cross-power spectrum, usually the real part or the magnitude, which represents the total energy of the coordinated movement of the device and the environment within the sensitive frequency band.

[0085] Calculate the deviation and total energy of the disturbance. This represents the average energy of two signals within the same frequency band, serving as a standardized benchmark;

[0086] F. Finally, calculate the average coupling strength factor. This factor It is a dimensionless normalized value, such as between 0 and 1, which accurately quantifies the coupling strength between equipment deviation and environmental disturbance in sensitive frequency bands.

[0087] like Figure 3 The key technical path of the spectrum coupling analysis step is demonstrated. This method is used to intelligently identify whether equipment malfunctions during dredging operations are affected by the coupling effect of environmental disturbances.

[0088] Figure 3 The red curve represents the spectral distribution of the equipment deviation signal, and the blue curve represents the spectral distribution of the environmental disturbance signal. It can be observed that in the low-frequency region, especially near 0.1 Hz, both curves exhibit significant amplitude peaks. This phenomenon of strong energy concentration in the same frequency band is an important characteristic for the system to determine resonance between the equipment and the environment. The system calculates the coupling strength factor based on this frequency overlap characteristic. When this factor exceeds a set ship-type-specific threshold, such as 0.7, it is determined to be a strong coupling state, and operational strategies such as reducing lateral speed or activating compensation devices are immediately triggered. Figure 3 The difference between the red and blue curves on the mid-frequency axis also represents the uncoupled frequency band, which helps the system eliminate false alarms caused by occasional fluctuations.

[0089] Ultimately, the system will continue to monitor. The value of . When If the coupling factor exceeds a threshold specifically set for that vessel type, such as 0.7 for the heave-torque coupling of a cutter suction jack, the system determines that a strong coupling anomaly has occurred. In this case, the system will immediately trigger vessel-specific countermeasures. For example, it may automatically activate the heave compensation system of the cutter bridge, or issue an alarm and suggestion to the operator to reduce the lateral movement speed to avoid the resonant frequency.

[0090] like Figure 4 The dynamic evolution curve of the average coupling strength factor is shown. This factor is a key quantitative indicator that reflects the degree of coupling between equipment deviation and environmental disturbance, obtained through spectral analysis.

[0091] exist Figure 4 In the diagram, the blue curve represents the fluctuation of the coupling strength calculated in real time during actual operation, exhibiting a certain periodic rise and fall pattern, which, combined with noise fluctuations, forms a complex change pattern. The red dashed line represents the preset coupling judgment threshold for this dredging vessel type, set at 0.7, which is the boundary standard used by the system to determine whether it is in a dangerous resonance state.

[0092] from Figure 4 As can be seen, the coupling factor crosses the threshold line and remains above it for several periods of time. For example, if the factor remains above the judgment value between 30 and 50 seconds, the system will immediately identify it as a strong coupling state and trigger specific countermeasures, such as reducing the cutter speed or activating the heave compensation mechanism. Conversely, when the curve is below the threshold, such as within the range of 0 to 20 seconds, the system will consider that there is no significant collaborative risk between the equipment and the environment, and normal operating strategies can be maintained.

[0093] Figure 4 The slow fluctuations and abrupt changes in the coupling strength also demonstrate the continuous sensing capability and sensitive response mechanism of the steps in this invention under dynamic environmental conditions.

[0094] Step S3: Generate dredging operation decision

[0095] Based on the results of model fusion and analysis in step S2, the decision generation module in this step will calculate and output a set of optimal operation parameters.

[0096] 1. Example of decision output:

[0097] For trailing suction hoppers: output optimal speed, rake head depth, mud pump power, and optimal overflow time, etc.

[0098] For cutter suction type: output optimal lateral speed, cutter speed, mud pump speed, trolley step distance, etc.

[0099] 2. Imposing ship-specific safety constraints: When generating the above optimal decisions, the system must consider safety as the highest priority constraint. Focusing solely on efficiency or energy consumption could lead to high-speed operations even in dangerous sea conditions, which is unacceptable.

[0100] This embodiment creatively uses the comprehensive wave hazard index, collected in step S1 and calculated in step S3, as a key safety constraint for decision generation. The calculation process of this index is as follows:

[0101] A. The system first utilizes the wave height extracted from the UAV image in step S1. ,cycle Crushing height Surge water level Surge duration Parameters such as these.

[0102] B. The system must also pre-configure a set of reference thresholds for the current ship type. These thresholds are type-specific because different sizes and types have completely different wave tolerances, including:

[0103] To preset reference wave height, This is a preset reference period. For example, for large trailing suction hopper dredgers, It could be set at 3.0 meters; for medium-sized cutter suction dredgers, It may only be 1.0 meter;

[0104] To preset the reference crushing height, To preset the reference fracture impact distance, Distance is affected by the breaking of waves;

[0105] To preset the reference surge level, To preset the reference surge duration, To preset the reference surge impact distance,

[0106] Distance affected by surge.

[0107] C. The system calculates the three sub-item hazard assessment indices respectively:

[0108] (1) Wave Hazard Assessment Index:

[0109]

[0110] This formula assesses the hazard of conventional wave height and period, where Indicates relative wave height. It represents the relative period. Generally, shorter period waves have a more severe impact on the hull, and the sum of the two is used as a measure of the danger.

[0111] The function is an sigmoid saturation function, whose output value is smoothly limited to the range of (-1, 1) or (0, 1). When the waves are much smaller than the reference value, the danger index is close to 0; when the waves are much larger than the reference value, the danger index approaches 1 (maximum danger), and does not increase indefinitely. This is consistent with reality, that is, the danger level will saturate.

[0112] (2) Wave breakage hazard assessment index:

[0113]

[0114] This formula assesses the hazard of breaking waves, which is particularly important when operating in shallow nearshore waters. The hazard is determined by the relative height of the breaking waves (…). ) and relative proximity ( (To be decided jointly)

[0115] and It is also an sigmoid saturation function, but its output range is strictly limited to (0, 1), and it has the same characteristics as... A similar advantage is that it provides a smooth, bounded risk assessment.

[0116] (3) Wave surge hazard assessment index:

[0117]

[0118] This formula assesses the hazard of surges, taking into account three factors: relative water level (…). ), relative duration ( ) and relative proximity ( ).

[0119] The Sigmoid function used is for the same reason as above, to ensure the non-linear saturation characteristics of risk assessment.

[0120] D. Finally, the system calculates the overall wave hazard index through weighted averaging. :

[0121]

[0122] The weighting coefficients of 0.4, 0.3, and 0.3 here were determined based on extensive historical data and expert experience. In this example, the risks of regular wave height and periodicity are considered to have the highest weight, while breakage and swell each account for 30%. These weights can also be dynamically adjusted according to the vessel type and operating area.

[0123] E. Application As a safety constraint:

[0124] Calculated A value between 0 and 1 is immediately sent to the decision generation module, which must satisfy the following condition when generating a decision: , The maximum tolerable hazard threshold specific to this ship type;

[0125] if If the warning threshold, such as 0.6, is exceeded, the decision-making module must automatically reduce its efficiency objective and increase its safety objective. For example, the generated decision will automatically include constraints such as reducing the speed by 30% and limiting mud power to 80%.

[0126] if If the stop threshold is exceeded, such as 0.85, the decision module will override all other optimization objectives and directly generate the highest priority instruction to immediately stop the operation and raise the rake arm / cutter head to a safe position on the water surface.

[0127] like Figure 5 The process of constructing the comprehensive wave hazard index and the index curve are shown. Figure 5 The blue curve represents the wave hazard index calculated based on wave height and period, reflecting the intensity change of the impact of traditional periodic waves on the hull. When the period rises at a high frequency, it means that the safety of operation decreases.

[0128] The green curve is the breakage danger index, which reflects the destructive impact of sudden changes in waves in shallow water. The index rises rapidly when the waves are steep and approach the preset warning range.

[0129] The cyan line represents the surge risk index, which is used to quantify the risk of non-periodic water level rise caused by storms or long-period waves. Its rise generally lags behind the normal wave height, but it has a significant impact under low-frequency strong wave conditions.

[0130] Figure 5 The most crucial element is the bold red curve, representing the comprehensive wave hazard index constructed in this invention. This curve integrates and weights the three sub-indicators mentioned above, forming a vital basis for the intelligent decision-making system to determine whether to adjust dredging operation parameters. When this index exceeds a specific threshold, for example in… Figure 5 If the index value reaches or exceeds 0.8 during multiple time periods, the system will immediately limit parameters such as rake head depth and mud pump power to ensure operational safety. Conversely, within a stable range where the index is below 0.4, the system can prioritize progress targets to improve efficiency.

[0131] Figure 5 The various curves fluctuate and intersect, but the composite curve maintains a continuous upward or downward trend, demonstrating the effectiveness and engineering practical value of the weighted construction mechanism in maintaining a stable response under multidimensional disturbance environments.

[0132] Step S4: Conduct ship-type-specific resilience monitoring of the effectiveness of dredging operation decisions.

[0133] In this step, the system must not only generate decisions, but also monitor the execution effect of the decisions in real time, and ensure that the system can remain resilient when encountering unexpected emergencies, that is, it can safely and controllably degrade and operate, rather than crashing directly.

[0134] The ship-specific resilience monitoring module in this embodiment is specifically designed to handle two typical emergency scenarios: environmental compliance and unknown geology.

[0135] 1. Triggering of Resilient Monitoring: The system monitors two high-level triggers in real time, including:

[0136] Trigger 1: Entering an environmentally sensitive area

[0137] Monitoring method: The system's decision server is pre-loaded with a GIS electronic map of the work area, which clearly marks environmentally sensitive areas, such as coral reef protection areas, important fish spawning areas, and areas near drinking water intakes;

[0138] Trigger: The system compares the real-time GNSS positioning of the vessel with the GIS map. Once it detects that the dredging vessel's operating radius or anchoring range has encroached on the electronic fence of the environmentally sensitive area, it will immediately trigger resilient monitoring.

[0139] Trigger 2: Encountering a Sudden Change in the Substrate

[0140] Monitoring method: Based on the analysis results of the dual-drive coupled sub-model in step S2;

[0141] Trigger: When the model detects a drastic and continuous anomaly in the ship type-specific data, such as the cutter torque suddenly soaring to its limit or the sludge discharge flow rate fluctuating at a persistently high level;

[0142] Distinguishing features: This trigger is different from spectral coupling, which may only be an efficiency issue, and also different from sensor failure, which is a data reliability issue; the trigger here clearly points to a sudden change in the physical world, that is, the discovery of hard rock, shipwreck, or highly viscous clay layers that were not shown in the exploration report.

[0143] 2. Execution of Resilient Monitoring: Once any of the above triggers is activated, the system will perform the following actions:

[0144] A. Combining real-time wave data: When executing monitoring actions, the system does not abruptly stop all operations, but also refers to the comprehensive wave hazard index calculated in step S3. This is a high-level security consideration.

[0145] For example, when encountering a sudden change in seabed sediment (cutting block jamming), if the sea conditions are also very bad... If the sea conditions are high, the system must adopt the most conservative strategy, such as immediately dumping soil and raising the bridge; if the sea conditions are good ( (Very low), the system can try more refined degradation operations, such as reversing the auger or reducing the speed.

[0146] B. Automatically generate ship-type specific downgrade coefficients: The system will generate one or a set of downgrade coefficients (multipliers between 0 and 1). These coefficients are ship-type specific and scenario-dependent.

[0147] C. Degradation-Based Operational Decisions: These generated degradation coefficients are immediately sent to the decision generation module in step S3. They will serve as the highest priority constraints, overriding the original optimal efficiency or energy consumption objectives. The decision module must, based on these degradation coefficients, regenerate and issue safety-degraded operational parameter instructions, such as instructions that limit torque and lateral speed.

[0148] 3. Regulatory Certificate: In step S4, the system will also generate a tamper-proof report specific to the vessel type. This report details when, where, why, how, and the result of the response. Utilizing cryptographic hashing or blockchain technology to prevent subsequent modification, this report is used for verification and auditing by the construction company, supervisor, and maritime authorities.

[0149] Step S5: Based on the above execution results, iteratively optimize the ship-specific dual-drive coupling sub-model.

[0150] Since dredging operations are not a one-time event but a long-term process, the system in this embodiment is designed to have online learning or reinforcement learning capabilities, specifically including the following processes:

[0151] 1. Feedback data source: The anti-tampering report generated in step S4 and the verified execution effect data;

[0152] 2. Optimization objective: The ship-specific dual-drive coupling sub-model in step S2;

[0153] 3. Optimization Process: For example, in step S4, the system records that the frequency of performing sediment degradation operations due to sediment mutation is very high in a specific area. The system's data-driven module will obtain these reports and learn that there is a strong correlation between the GIS coordinates of a specific area and features such as high torque and high sludge discharge pressure fluctuations. In the next iteration of optimization, the weights of the neural network model will be updated.

[0154] 4. Effect: When the dredging vessel is about to operate near the GIS coordinates of a specific area, the model in step S2 can predict in advance that the bottom sediment is hard. Therefore, in step S3, the system will no longer blindly pursue high efficiency, but will proactively reduce the cutter speed and lateral movement speed, thereby avoiding the passive triggering of degradation operations and realizing an intelligent upgrade from passive resilience monitoring to proactive predictive decision-making.

[0155] This first embodiment not only resolves the contradiction between vessel versatility and precision, but also, through refined environmental risk quantification, in-depth operational condition coupling diagnosis, and robust resilience monitoring strategies, enables the entire dredging operation decision-making system to exhibit unprecedented levels of safety, adaptability, and intelligence when facing complex, dynamic, and high-risk real marine environments.

[0156] Example 2:

[0157] This second embodiment further describes how to specifically implement the construction details of the model and its core robust self-diagnostic mechanism in step S2 of embodiment one: constructing and executing the ship-type-specific dual-drive coupling sub-model.

[0158] This embodiment aims to solve the prominent problem of the conflict between adaptability and stability in the background technology, especially the technical problem that when the model encounters data pollution such as sensor failure or extreme working conditions, the system incorrectly adjusts the model parameters, leading to model collapse or decision-making errors.

[0159] I. Architecture and Operation of the Dual-Drive Coupled Sub-Model

[0160] The ship-specific dual-drive coupled sub-model used in this embodiment is an advanced hybrid modeling method. Its core lies in combining rigorous physical mechanisms with flexible data-driven capabilities, and realizing mutual supervision and correction between the two.

[0161] 1. Model Components: This model must include at least the following two levels:

[0162] Mechanism-driven layer: This layer consists of mathematical models built upon the principles of physics, fluid dynamics, and engineering mechanics specific to certain dredging vessel types; in other words, dredging-specific physical models. For example, for cutter suction dredging vessels, this layer includes energy loss equations for sediment two-phase flow in pipelines, soil resistance calculation formulas for the cutter head, and force models of the lateral movement system on the hull. The main function of the mechanism-driven layer is to provide accurate physical constraints and interpretable baseline predictions.

[0163] Data-driven layer: This layer is a black-box model composed of algorithms such as deep neural networks, support vector machines, or reinforcement learning. It is trained using a large amount of historical operational data and real-time sensor data to capture complex, nonlinear, and high-dimensional relationships that are difficult for mechanistic models to describe precisely. For example, the nonlinear relationship between mud pump speed and actual conveying efficiency in a specific sediment environment.

[0164] 2. Model Coupling Mechanism: During model operation, the mechanism-driven layer and the data-driven layer do not run in isolation, but are dynamically coupled and mutually corrected through a bias analyzer. This coupling mechanism is reflected in the following aspects:

[0165] First, the mechanism-driven layer provides a mechanism prediction value based on the current operating parameters. For example, predicting pipeline pressure at the current flow rate;

[0166] Secondly, the data-driven layer provides a predicted data value based on real-time sensor input. ;

[0167] Next, the system directly reads the actual measured values ​​from the field sensors. The system uses mechanisms to predict values. Forecast values ​​of the data Apply constraints and calculate the deviations between them. This deviation Defined as the difference between the predicted value and the actual measured value, i.e.:

[0168]

[0169] This deviation This represents the inherent error of the mechanistic model and the degree of deviation between real-time data and physical principles. The core of the dual-drive coupling mechanism lies in the fact that the model will adjust based on this deviation. The size of, and a preset first deviation threshold. This determines whether the current model parameters need adjustment, thereby achieving adaptive iterative optimization of the model, including:

[0170] when When the condition is within a certain range, it indicates that the mechanism model fits the actual operating conditions well, or the deviation is within an acceptable range. At this time, the model maintains its current parameters.

[0171] when This indicates a significant deviation between the model and reality. At this point, the system triggers an adaptive mechanism to fine-tune some parameters in the data-driven or mechanism-driven layer. For example, it might update the neural network weights based on gradient descent or correct the friction coefficient in the mechanism model to reduce... This allows the model to adapt to new working conditions or environments.

[0172] like Figure 6 The key comparison logic in the real-time execution process of the dual-drive coupled sub-model is demonstrated. It generates a deviation value by simultaneously monitoring the difference between the physical model output and the measured value of the device sensor to determine whether the model deviates from the actual working condition.

[0173] exist Figure 6In the diagram, the blue curve represents the predicted value output by the mechanism-driven model, while the green curve represents the measured value collected by the system through field sensors. The two curves show similar trends for most of the time, indicating that the model performs well under stable conditions. However, during certain periods, particularly between 30 and 45 seconds, the green curve exhibits increased fluctuations and a slight phase shift, significantly widening the difference from the blue curve. The red curve on the right is the deviation curve constructed based on the difference between the two curves. It reflects the degree of fit between the model and the field data. When the deviation value continues to rise and exceeds the system's set judgment threshold, the system will trigger a fusion mechanism to diagnose whether it is model drift or sensor anomaly. If the deviation anomaly is identified as data contamination, the system will pause model parameter updates and activate emergency strategies to prevent the model from being misled by erroneous data.

[0174] II. Diagnosis of Fusion Mechanism

[0175] While the above coupling mechanism achieves adaptability, it has a fatal flaw: when the sensor malfunctions, such as a pressure sensor reading of zero or at its maximum output, It can suddenly become very large, exceeding The adaptive mechanism may mistakenly interpret this as a defect in the model itself, and thus frantically adjust the model parameters, ultimately leading to model contamination or even collapse.

[0176] This embodiment introduces a highly robust self-diagnosis and triage mechanism for this conflict problem, namely the fusion mechanism diagnosis. This mechanism introduces two additional technical steps on the basis of the original: a second deviation threshold (mutation deviation threshold) and deviation continuity monitoring, to distinguish between two fundamentally different anomaly types: model drift and data pollution.

[0177] 1. Introduce dual thresholds and deviation continuity monitoring: Based on the original first deviation threshold... Based on this, this embodiment adds two key monitoring parameters:

[0178] (1) Second deviation threshold Set a value much larger than A threshold is used to determine whether an extreme anomaly has occurred. For example, It could be 5% of the predicted value, while That would be 30% of the predicted value;

[0179] (2) Judgment of deviation continuity Set a minimum duration to distinguish between transient noise and persistent anomalies. For example, It can be set to 5 seconds.

[0180] 2. Mechanism Logic and Triage Processing: The system will continuously monitor... The results are then processed in three ways:

[0181] Scenario 1: Normal operation or model fine-tuning, i.e. Smaller, not exceeding ,at this time:

[0182] state: ;

[0183] Diagnostic conclusion: The model basically matches the actual working conditions;

[0184] Solution: Maintain existing parameters or perform minimal background optimization.

[0185] Scenario 2: Model drift triggers adaptive adjustment, i.e. Larger, but short-lived or within an acceptable range, in this case:

[0186] state: And the duration is less than ;

[0187] Diagnostic conclusion: It is believed that a slow change in the environment or substrate has caused the model predictions to drift from reality;

[0188] Handling measures: Trigger the adaptive iterative optimization in the first part of Implementation Example 2. For example, the system will be based on... Retrain the data-driven layer model parameters to realign the model with actual operating conditions. This adjustment is gradual and constrained, and will not cause the model to run away with the flow.

[0189] Scenario 3: Data contamination or extreme operating conditions trigger resilience measures, i.e. Enormous and lasting for a long time, at this point:

[0190] state: ,or Persistently greater than And the duration exceeds ;

[0191] Diagnostic conclusion: This is the highest level of anomaly. The system immediately determined that a sensor malfunction, communication interruption, or extreme external interference had occurred, indicating an unreliable anomaly in the data source or physical conditions themselves.

[0192] Mitigation measures: This is the most critical aspect of this embodiment, ensuring the robustness of the model, and includes the following:

[0193] (1) Isolate abnormal data sources: The system will generate sensor data with huge deviations, which will lead to The data is marked as unreliable or faulty, and this isolated data is completely shielded or not adopted in subsequent decision generation and model optimization processes;

[0194] (2) Switch to Emergency Model: The system's decision generation module immediately invokes a preset, static emergency model or safety model. This emergency model no longer relies on current real-time anomaly data, but instead makes conservative decisions based on historical safety data and an expert rule base. For example, the sludge pump speed is forcibly locked at a safe value, and the cutter torque is limited to a safe upper limit.

[0195] (3) Send alarms: The system sends high-level sensor failure or extreme working condition alarms to operators and remote monitoring centers, and prompts the affected sensors or model components, requiring manual intervention for verification.

[0196] 3. Automatic model recovery and takeover: The system continues to monitor in the background during the execution of the emergency model. Once the isolated sensors return to normal, or the extreme external interference disappears, leading to... Falling back down After a period of time, the system will determine that the abnormal operating condition has been resolved. At this point, the system will smoothly switch back from the emergency model to the adaptive model, restoring the dredging operation to an intelligent optimization state.

[0197] The dual-drive coupling model and self-diagnostic mechanism proposed in this second embodiment introduce a dual-bias threshold. and They successfully distinguished technically between two completely different types of substances that both lead to high... Anomalies: Model drift (correctable) and data contamination (requires isolation).

[0198] Adaptability to model drift: through This ensures that the model can continuously and gradually adjust itself as parameters such as substrate and environment change slowly, maintaining the system's adaptability and high efficiency.

[0199] Robustness to data contamination: through This ensures that in the event of sudden high-deviation events such as sensor failure, the system can immediately perform data isolation and emergency takeover, effectively preventing erroneous data from contaminating or poisoning the core model. This completely solves the technical problem of adaptive systems being prone to collapse under harsh working conditions in the background technology, and greatly improves the reliability and safety of the dredging operation decision system.

[0200] In summary, the technical solution of Embodiment 2 endows the entire system with a high degree of self-diagnosis capability and resilience for fault-tolerant operation in a dynamic and uncertain marine environment.

[0201] Example 3:

[0202] In the background technology, intelligent decision-making systems face a serious technical problem of conflict between responsiveness and stability: on the one hand, the system needs to respond sensitively to changes in the external environment to ensure safety; on the other hand, if environmental signals, such as the wave hazard index, fluctuate or oscillate at high frequency around a certain decision threshold, an overly sensitive system may lead to catastrophic high-frequency switching of control commands, causing severe equipment wear or control failure.

[0203] The purpose of this embodiment is to completely resolve the aforementioned conflict through a sophisticated weight adjustment strategy, enabling the system to respond sensitively to real-world environmental changes while remaining stable against meaningless noise fluctuations, and ensuring a smooth and shock-free process when switching strategies is necessary. Specifically, this includes the following:

[0204] I. Establishing the Relationship between Objective Function and Weights

[0205] During dredging operations, the decision-making system always faces at least two conflicting core optimization objectives: one is the energy consumption objective, and the other is the schedule objective.

[0206] 1. Definition of objective function: The system internally defines at least two optimization objective functions:

[0207] Energy consumption objective function This function aims to minimize the unit cost of the job. It is a composite function that includes: the total power consumption of key equipment such as the main mud pump, flushing pump, cutter head, and transverse winch of the dredging vessel (such as a cutter suction dredger); it can also be the fuel consumption per unit time or per unit volume of earthwork; in some cases, it can also include the wear and depreciation model of the equipment, because high energy consumption operation is usually accompanied by high wear.

[0208] Schedule objective function This function aims to maximize the efficiency of the job. It is also a composite function, which includes: the volume of excavated earth per unit time. For example, it can be calculated by multiplying the flow rate and concentration through the sludge discharge pipe; for trailing suction hopper dredgers, it can be the loading efficiency per scoop or the number of daily cycles.

[0209] These two goals are physically inherently conflicting. For example, in order to improve... Operators typically increase the mud pump speed and cutter power, but this inevitably leads to... A sharp rise.

[0210] 2. Introduction of weighting relationship: In order to find an optimal balance between the two conflicting objectives of energy consumption and schedule, this embodiment introduces dynamic weights.

[0211] The final comprehensive objective function of the system optimization It is a weighted sum of these two sub-objective functions, with the system having an energy consumption objective function. An energy consumption weight was assigned. and the schedule objective function A progress weight was assigned. At any given time, these two weights must satisfy a normalization constraint, namely:

[0212]

[0213] This constraint indicates that: if If set to 0.8, then It must be set to 0.2 automatically. This indicates that the system's highest priority is currently energy conservation and safety, as high energy consumption is often accompanied by high risk, while progress is secondary. Conversely, if If set to 0.9, then It must be 0.1 to indicate that the system is in a rush mode.

[0214] 3. Introduction of triggering factors: In traditional static schemes, and It may be set to a fixed value before the start of the entire construction phase, but this method cannot cope with dynamic marine environments.

[0215] A key point of this invention is that, and The value is dynamically adjusted. The system introduces a trigger factor, which is used to determine the value in real time. and The specific value.

[0216] Optionally, the triggering factor is directly derived from the comprehensive wave hazard index calculated in Example 1. , The degree of danger in the current working environment has been quantified. With weight , The logical relationship between them is clear:

[0217] when At extremely low levels (calm and peaceful), environmental risk is minimal. In this situation, the system should boldly pursue efficiency. (Schedule weight) is set to a high value, (Energy consumption / safety weight) is set to a low value;

[0218] when Extremely high sea conditions (severe sea states) mean extremely high environmental risks. In this situation, safety must be the top priority, and the system must be forced to... Set to a high value. For example, reducing energy consumption can decrease equipment load and risk, while It must be set to a low value.

[0219] II. Solving the jitter problem: High-stability hysteresis and smooth transition mechanism

[0220] Will Adjust as a trigger factor and While achieving responsiveness, this also introduces the oscillation problem mentioned earlier. If If the calculated value fluctuates around a set single-point threshold, such as 0.6 (0.59, 0.61, 0.58, 0.62…), due to sensor noise or instantaneous fluctuations in ocean waves, the system will then command… and The system oscillates wildly between high and low values. This high-frequency weight oscillation directly translates into control command oscillations, which no physical system can withstand, causing enormous mechanical shocks and safety hazards.

[0221] To completely resolve this conflicting technical problem, this invention introduces a dual high-stability mechanism consisting of a hysteresis threshold and a smooth transition.

[0222] Mechanism 1: Hysteresis Control Based on Dual Thresholds

[0223] This embodiment does not use a single decision threshold, but sets two different thresholds, namely:

[0224] Open threshold This is a high dangerous activation threshold;

[0225] Close threshold This is a low danger removal threshold.

[0226] These two thresholds must satisfy numerically The relationship.

[0227] For example, in this embodiment, It was set to 0.7, while The threshold is set to 0.5. Through these two thresholds, the system defines three distinct state ranges and switching logic:

[0228] Status 1: Safety and Schedule Priority Zone ( );

[0229] Scenario: When A value below 0.5 indicates that the sea state is very safe;

[0230] Action: The system will (Schedule weight) is set to a high value. For example... .

[0231] Status 2: Danger and Safety Priority Zone ( );

[0232] Scenario: When A value above 0.7 indicates that the sea state has entered a dangerous level;

[0233] Action: The system will (Energy consumption / safety weight) is set to a high value. For example... .

[0234] State 3: Hysteresis Holding Zone ( );

[0235] Scenario: When When it falls exactly between 0.5 and 0.7;

[0236] Action: The system does not perform any new switching action, but maintains the state before the switch.

[0237] If the system is from the safe zone ( If the value rises to 0.6, the system will maintain a progress-priority state. );

[0238] If the system is from the danger zone ( If the value drops to 0.6, the system will maintain a safety-first state. ).

[0239] Regarding how the hysteresis mechanism solves the jitter problem, we can simulate a jitter scenario here:

[0240] It fluctuates around 0.6, for example, 0.59, 0.61, 0.58, 0.62… because all these jittery values ​​(0.58 to 0.62) fall within… (0.5) and Within the hysteresis hold-up region between (0.7); therefore, regardless of Within this range, regardless of high-frequency oscillations, the system will ignore this noise and maintain its state before entering that region, whether prioritizing safety or progress. Therefore, the weights... and It will remain stable, without any switching, and the oscillation of control commands will be completely eliminated.

[0241] Only when the waves are big enough to truly cross The system will only switch to safety priority when the wave size is (0.7); and only when the wave size is truly below the required level. The system will only switch back to progress priority when the threshold is 0.5. This hysteresis loop design perfectly balances responsiveness to real signals and stability against noisy signals.

[0242] Mechanism 2: Smooth Transition Function

[0243] The aforementioned hysteresis mechanism addresses the question of when to switch, but not how to switch. If the system... In an instant, The jump from 0.1 to 0.8 is mathematically a step change. This kind of step-like weight change also causes abrupt changes in control commands. Such a sudden and significant change can have a powerful impact on the mechanical system, which is something that should be avoided in engineering. To address this impact problem, this embodiment introduces a smooth transition function.

[0244] The smooth transition function ensures that when the system decides to switch weights, i.e., when it crosses... or At that time, weight and The changes are gradual and smooth, rather than instantaneous.

[0245] For example, when the system detects Crossed (0.7), needs to be When switching from the current value of 0.1 to the target value of 0.8, the system will not execute immediately. ;

[0246] Instead, the system will start a transition timer and gradually change the time over a preset period using a function. For example, through:

[0247] Linear transition: The value increases linearly from 0.1 to 0.8 within 10 seconds. For example, it increases by [value] per second. ;

[0248] S-shaped function transition: using In the form of, It is an S-shaped function that smoothly changes between 0 and 1. The advantage of the S-shaped function is that its rate of change (derivative) is close to zero at the beginning and end of the change, which makes the entire transition process the gentlest at both ends and the fastest in the middle, achieving the best smoothness and minimizing the impact on physical devices.

[0249] Based on the above implementation, Throughout the smooth increase, It will be reduced smoothly and synchronously to meet the requirements at all times. Constraints. For example, in seconds, The transition was smooth to 0.45, then... It must be exactly equal to that moment. .

[0250] This embodiment is illustrated by... The weighting relationship, and the weighting relationship between them. The triggering factors constitute the basis for adaptive decision-making; more importantly, it establishes the basis for decision-making through hysteresis double thresholds. / This mechanism fundamentally solves the stability problem of oscillation control; simultaneously, through a smooth transition function mechanism, it addresses the smoothness problem of switching shocks. Ultimately, in this embodiment, the system achieves a high-level decision-making system capable of sensitively responding to real dangers, intelligently filtering false noise, and smoothly protecting physical equipment. It perfectly resolves the conflict between responsiveness and stability in the background technology, enabling dredging operations to achieve crucial control stability and mechanical safety on the path to intelligent and unmanned operation.

[0251] like Figure 7 A simulation diagram of the smooth transition function is shown. Figure 7 The upper-middle section displays the dynamic changes of energy consumption weight and schedule weight, with the red curve representing energy consumption weight and the blue curve representing schedule weight. The changes of the two curves strictly adhere to the constraint that the sum of the weights must always be one. The lower section displays the changes of the triggering factor, namely the comprehensive wave hazard index, as well as the hysteresis interval formed by the set opening and closing thresholds.

[0252] from Figure 7 It can be clearly observed that when the trigger factor value crosses the opening threshold (represented by the red dashed line) upwards from the safe zone, the system begins a smooth transition from the progress-first mode to the safety-first mode. At this point, the energy consumption weight gradually increases from a lower level while the progress weight decreases accordingly. The entire transition process continues for a preset time period, and the change curve exhibits an S-shaped characteristic, ensuring that the rate of change at the beginning and end of the weight adjustment is minimized, effectively avoiding equipment shock problems caused by sudden weight changes in traditional systems. Similarly, when the trigger factor value crosses the closing threshold (represented by the blue dashed line) downwards from the danger zone, the system smoothly reverts to the progress-first mode.

[0253] Of particular note is that within the hysteresis hold-off zone where the trigger factor value is between the two thresholds, the system weights remain stable despite normal fluctuations in environmental parameters. This demonstrates that the hysteresis control mechanism employed in this invention can effectively filter environmental noise fluctuations and prevent harmful oscillations in control commands near high-risk thresholds. The entire weight adjustment process ensures both the system's sensitive response to real-world environmental hazards and the stability and continuity of control commands, solving the technical challenge of balancing responsiveness and stability. This provides a reliable decision-making basis for the intelligent control of dredging operations.

[0254] Example 4:

[0255] like Figure 8 As shown in the above method embodiments, the present invention also proposes an intelligent decision-making optimization system for dredging operations based on multi-source data fusion, comprising:

[0256] The data acquisition module is used to collect and preprocess multi-source data, which includes general data and ship-specific data for different dredging vessel types.

[0257] The model execution module is used to construct and execute a ship-specific dual-drive coupling sub-model based on the dredging vessel type, fuse the multi-source data, and perform ship-specific spectral coupling analysis and ship-specific active anomaly identification.

[0258] The decision generation module is used to perform ship-type-specific dynamic weight adjustment and ship-type-specific safety constraints based on the results of the multi-source data fusion, and call the ship-type-specific pre-trained working condition library to generate dredging operation decisions.

[0259] The monitoring and verification module is used to monitor and verify the execution effect of the dredging operation decision on a ship type-specific resilience basis, and generate a ship type-specific anti-tampering report.

[0260] The iterative optimization module is used to iteratively optimize the ship-type-specific dual-drive coupling sub-model based on the execution results.

[0261] This system combines a unified framework with ship-specific features by constructing a dual-drive coupled sub-model and data system specific to each ship type. This allows the decision-making system to accurately match the unique physical mechanisms of a specific ship type while operating within a unified intelligent framework. In particular, by introducing a strategy with hysteresis and smooth transition, the system achieves a sensitive response to environmental changes while eliminating control command oscillations caused by data noise jitter to a certain extent. This balances the responsiveness and stability of the system, ensuring the smooth operation of the physical equipment.

[0262] Example 5:

[0263] Corresponding to the above embodiments, the present invention also proposes an electronic device.

[0264] like Figure 9 The diagram shows a structural schematic of an electronic device according to the present invention. The electronic device 100 includes a processor 101 and a memory 103. The processor 101 and the memory 103 are connected, for example, via a bus 102. Optionally, the electronic device 100 may further include a transceiver 104. It should be noted that in practical applications, the transceiver 104 is not limited to one unit, and the structure of this electronic device 100 does not constitute a limitation on the embodiments of the present invention.

[0265] Processor 101 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in connection with this disclosure. Processor 101 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0266] Bus 102 may include a pathway for transmitting information between the aforementioned components. Bus 102 may be a PCI bus or an EISA bus, etc. Bus 102 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0267] The memory 103 stores a computer program corresponding to the intelligent decision-making optimization method for dredging operations based on multi-source data fusion according to the above embodiments of the present invention. This computer program is controlled and executed by the processor 101. The processor 101 executes the computer program stored in the memory 103 to implement the content shown in the aforementioned method embodiments.

[0268] Among them, electronic devices 100 include, but are not limited to: mobile terminals such as laptops and PADs (tablet computers) and fixed terminals such as desktop computers. Figure 9 The electronic device 100 shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0269] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for intelligent decision optimization of dredging operation based on multi-source data fusion, characterized in that, The method comprises the following steps: Collecting and pre-processing multi-source data, the multi-source data including general data and ship type specific data for different dredging ship types; the general data including sea wave data; the sea wave data including wave height extracted from sea wave images collected by a drone , period , sea wave breaking height , surge water level and surge duration ; the dredging ship types including a drag-suction dredger, a cutter-suction dredger and a grab dredger; Based on the dredger type, a ship type specific dual drive coupling sub-model is constructed and executed, the multi-source data is fused, and ship type specific spectrum coupling analysis and ship type specific active anomaly identification are performed; Based on the results of the multi-source data fusion, ship type specific dynamic weight adjustment and ship type specific safety constraints are performed, and a ship type specific pre-trained working condition library is called to generate a dredging operation decision; The execution effect of the dredging operation decision is subjected to ship type specific resilience supervision and verification, and a ship type specific tamper-proof report is generated; Based on the execution effect, the ship type specific dual drive coupling sub-model is iteratively optimized; The construction and execution of the ship type specific dual drive coupling sub-model comprises: Constructing a mechanism layer: based on a dredging specific physical model, the dredging specific physical model is used to describe the sediment transport, fluid dynamics and energy consumption characteristics of a specific dredger type, and ship type specific correction terms are introduced for the dredger type; Constructing a data layer: extracting and fusing the ship type specific data and the general data with strong correlation with the dredging performance indicators, the strong correlation characteristic parameters include wave height and hopper loading capacity; Establishing a fusion mechanism: calculating the deviation of the mechanism layer output value and the data layer feature vector, and automatically adjusting the preset parameters in the dredging specific physical model when the deviation is greater than the preset deviation threshold; The fusion mechanism further comprises: Set a mutation deviation threshold greater than the preset deviation threshold; When the deviation is greater than the mutation deviation threshold, or the deviation is still greater than the preset deviation threshold after one or more automatic adjustments, suspend the automatic adjustment and trigger the fusion mechanism diagnosis; The fusion mechanism diagnosis includes: consistency check on the multi-source data constituting the data layer feature vector; If the consistency check determines that there is an abnormal data source, isolate the abnormal data source, enable the data reconstruction model to generate replacement data for the deviation calculation, and restore the automatic adjustment; If the consistency check determines that the data sources are consistent, it is determined that the dredging specific physical model is invalid, the automatic adjustment is frozen, and an emergency response model is called from the ship type specific pre-trained working condition library to replace the dredging specific physical model.

2. The method of claim 1, wherein, The ship type specific data comprises: For the drag suction dredger, the ship type specific data at least includes hopper loading capacity and hopper sediment sloshing amplitude; For the cutter suction dredger, the ship type specific data at least includes the pressure of the sludge discharge pipe and the sludge flow fluctuation value; For the grab dredger, the ship type specific data at least includes the grab opening degree and the grab lifting speed.

3. The method of claim 1, wherein, The method also comprises calculating a sea danger index based on said sea data and for the ship type specific safety constraints; The sea wave integrated risk index ; wherein is a sea wave danger evaluation index, is a sea wave breaking danger evaluation index, is a sea wave surge danger evaluation index.

4. The method of claim 3, wherein, The sea wave danger evaluation index wherein is a preset reference wave height, is a preset reference period; The sea wave breaking danger evaluation index wherein is a preset reference breaking height, is a sea wave breaking influence distance, is a preset reference breaking influence distance, is a natural constant; The sea wave surge danger evaluation index wherein is a preset reference surge water level, is a preset reference surge duration, is a surge influence distance, is a preset reference surge influence distance.

5. The method of claim 1, wherein, The ship type specific spectrum coupling analysis comprises: For each dredging vessel type, its sensitive frequency band is adapted; the principal component sequence of equipment deviation and the environmental disturbance sequence are obtained, and the deviation power spectrum is calculated respectively. Disturbance power spectrum and cross-power spectrum ; Within the sensitive frequency band, a low frequency coherency measure is calculated , and the deviation and the total perturbation energy ​ calculating an average coupling strength factor and when the average coupling strength factor is greater than a ship-specific coupling factor threshold, determining strong coupling and performing ship-specific countermeasures.

6. The method of claim 3, wherein, The ship type specific dynamic weight adjustment comprises: Setting energy consumption weight With progress weight Wherein ; For the dredger type, a ship type specific trigger factor is set by combining parameters in the ship type specific data and the sea wave data; dynamically adjust the weight distribution of and when the ship type specific trigger is triggered.

7. The method of claim 6, wherein, Set a start threshold and an exit threshold for the ship type specific trigger factor, the exit threshold is lower than the start threshold; The dynamic adjustment And The weight distribution, specifically including: triggering the weight distribution to adjust to a first target weight group when the value of the ship-type-specific trigger factor rises and is greater than the start threshold value; triggering the weight distribution to adjust to a second target weight group when the value of the ship-type-specific trigger factor falls and is less than the exit threshold value; And at the time of triggering the weight distribution adjustment, the And The values of the first and second target weight sets are based on a preset smooth transition function, and gradually change to the first or second target weight set within a preset transition time period to avoid job state oscillation caused by weight mutation.

8. The method of claim 3, wherein, the ship-type-specific resilience management comprises: for the dredging ship type, when the dredging ship is monitored to enter an environmentally sensitive area or encounter a bottom condition mutation scenario; in combination with the real-time sea wave data, automatically generating a ship-type-specific operation parameter degradation coefficient; and adjusting the operation parameters in the dredging operation decision based on the operation parameter degradation coefficient.

Images